blob_id
stringlengths
40
40
directory_id
stringlengths
40
40
path
stringlengths
2
327
content_id
stringlengths
40
40
detected_licenses
listlengths
0
91
license_type
stringclasses
2 values
repo_name
stringlengths
5
134
snapshot_id
stringlengths
40
40
revision_id
stringlengths
40
40
branch_name
stringclasses
46 values
visit_date
timestamp[us]date
2016-08-02 22:44:29
2023-09-06 08:39:28
revision_date
timestamp[us]date
1977-08-08 00:00:00
2023-09-05 12:13:49
committer_date
timestamp[us]date
1977-08-08 00:00:00
2023-09-05 12:13:49
github_id
int64
19.4k
671M
star_events_count
int64
0
40k
fork_events_count
int64
0
32.4k
gha_license_id
stringclasses
14 values
gha_event_created_at
timestamp[us]date
2012-06-21 16:39:19
2023-09-14 21:52:42
gha_created_at
timestamp[us]date
2008-05-25 01:21:32
2023-06-28 13:19:12
gha_language
stringclasses
60 values
src_encoding
stringclasses
24 values
language
stringclasses
1 value
is_vendor
bool
2 classes
is_generated
bool
2 classes
length_bytes
int64
7
9.18M
extension
stringclasses
20 values
filename
stringlengths
1
141
content
stringlengths
7
9.18M
6ed9bc3b190de6a63010d8ed338a5446d33ad24f
c7825ebaf6170b33d232afec734edb83b23cc2d1
/origin_datas/TCGA/0_datadeal.R
4e675ad31253b433202a70fb36aa7415fb03756d
[]
no_license
yancunlibmchh/TNBC_hypoxia
79e82d9c210439729a8735eb94b440d875130ac1
8d2959cce129dca0f8ec0f5608a3aa75257e892c
refs/heads/master
2023-06-25T14:50:24.412497
2021-07-30T09:19:37
2021-07-30T09:19:37
390,946,198
0
4
null
null
null
null
UTF-8
R
false
false
2,819
r
0_datadeal.R
source("http://bioconductor.org/biocLite.R") options(BioC_mirror="http://mirrors.ustc.edu.cn/bioc/") biocLite("genefu") biocLite("amap") biocLite("breastCancerMAINZ") biocLite("breastCancerTRANSBIG") install.packages('R.utils') install.packages('xtable') install.packages('caret') library(amap) library(genefu) library(xtable) library(caret) library(breastCancerMAINZ) library(breastCancerTRANSBIG) library(data.table) library(R.utils) setwd("E:/0_test/01_TNBC/0_data") mRNA_exp_count <- fread("TCGA-BRCA_mRNA_counts.txt.gz",header = T,quote = "",sep="\t") mRNA_exp_count <- as.data.frame(mRNA_exp_count) miRNA_exp_count <- fread("TCGA-BRCA_counts.txt.gz",header = T,quote = "",sep="\t") miRNA_exp_count <- as.data.frame(miRNA_exp_count) sample_info <- fread("BRCA_survival.txt.gz",header = T,quote = "",sep="\t") length(intersect(as.matrix(sample_info)[,1],colnames(mRNA_exp_count))) sample_info_more <- fread("TCGA-BRCA_clinical.csv",header = T,quote = "",sep=",") length(intersect(as.matrix(sample_info_more)[,1],substr(colnames(mRNA_exp_count),1,12))) ####### genefu pam50 ####### exp_mRNA_count <- mRNA_exp_count[,-1] rownames(exp_mRNA_count) <- mRNA_exp_count[,1] labels_sample <- as.numeric(substr(colnames(exp_mRNA_count),14,15)) samples_norm <- colnames(exp_mRNA_count)[which(labels_sample>=10)] samples_dis <- colnames(exp_mRNA_count)[which(labels_sample<10)] IDlist <- as.matrix(read.table("gene_ENSG2Symbol2ID.txt",header = F,quote = "",sep="\t")) interID <- intersect(IDlist[,1],rownames(exp_mRNA_count)) rownames(IDlist) <- IDlist[,1] EntrezGene.ID <- IDlist[interID,3] ind_NA <- which(is.na(EntrezGene.ID)) interID_noNA <- interID[-ind_NA] EntrezGene.ID_noNA <- EntrezGene.ID[-ind_NA] ID_probe <- as.matrix(read.table("gene_probe.txt",header = F,quote = "",sep="\t",fill = T)) interID_probe <- intersect(interID_noNA,ID_probe[,4]) ID_probe_useful <- ID_probe[which(ID_probe[,4]%in%interID_probe),] rownames(ID_probe_useful) <- ID_probe_useful[,4] probes <- ID_probe_useful[interID_probe,1] EntrezGene.ID_noNA_probe <- as.numeric(as.character(EntrezGene.ID_noNA[interID_probe])) ddata <- t(exp_mRNA_count[interID_probe,samples_dis]) dannot <- data.frame(probe=probes,Gene.symbol=interID_probe,EntrezGene.ID=EntrezGene.ID_noNA_probe) colnames(ddata) <- dannot$probe PAM50Preds<-molecular.subtyping(sbt.model = "pam50",data=ddata, annot=dannot,do.mapping=TRUE) sample_class <- data.frame(samples=names(PAM50Preds$subtype),class=as.character(PAM50Preds$subtype)) sample_class <- rbind(as.matrix(sample_class),cbind(samples_norm,"norm")) write.table(sample_class,"sample_class.txt",quote=F,row.names=F,col.names=T,sep = "\t") setwd("E:\\0_test\\01_TNBC") save.image(file = "allInfo.RData") load("allInfo.RData")
aaf7f30422db84c6b0664bfd7c61b63680287f53
013276f4e5be344d9ad7f21bfa2257338b43aa1b
/tests/testthat.R
4ed0b3e5469738d7765c5aca1d699ecbdf1e98b8
[ "MIT" ]
permissive
jimsforks/taskdesignr
a9e934ce85257cea22566cf589fd64aa5db59aed
01940d020b23658232dfd6d4c2791c444833f8a2
refs/heads/master
2023-02-23T06:55:07.138703
2021-01-31T00:53:47
2021-01-31T00:53:47
null
0
0
null
null
null
null
UTF-8
R
false
false
66
r
testthat.R
library(testthat) library(taskdesignr) test_check("taskdesignr")
db5278cbb09d456598e0cc9434bcbe4161dc8c0d
2a7e77565c33e6b5d92ce6702b4a5fd96f80d7d0
/fuzzedpackages/DiffusionRjgqd/R/JGQD.dic.R
7c910d0f1c521b6da7d493e13ea3f46b823a928c
[]
no_license
akhikolla/testpackages
62ccaeed866e2194652b65e7360987b3b20df7e7
01259c3543febc89955ea5b79f3a08d3afe57e95
refs/heads/master
2023-02-18T03:50:28.288006
2021-01-18T13:23:32
2021-01-18T13:23:32
329,981,898
7
1
null
null
null
null
UTF-8
R
false
false
1,330
r
JGQD.dic.R
JGQD.dic <-function(model.list,type='col') { M=matrix(0,6,length(model.list)) for(i in 1:length(model.list)) { M[,i]=unlist(model.list[[i]]$model.info)[1:6] } whDIC = which(as.numeric(M[4,])==min(as.numeric(M[4,]))) probs=as.numeric(M[4,whDIC])-as.numeric(M[4,]) M[c(3,4,5),]=format(round(as.numeric(M[c(3,4,5),]),2),nsmall=2) wh=which(as.numeric(M[4,])==min(as.numeric(M[4,]))) M[4,wh]=paste0(' [=] ',format(as.numeric(M[4,wh]),nsmall=2)) rownames(M)=c('Elapsed time :' ,'Time Homogeneous :' ,'p :' ,'DIC :' ,'pD (effective) :' ,'N :') mtags=rep(0,length(model.list)) for(i in 1:length(model.list)) { mtags[i] = model.list[[i]]$model.info$Tag } colnames(M) = mtags if(all(is.na(mtags))) { colnames(M)=paste('Model',1:length(model.list)) #warning('Some model tags are NULL!') } if(type=='row') { return(data.frame(M)) } if(type=='col') { M=t(M) colnames(M)=c('Elapsed_Time' ,'Time_Homogeneous' ,'p' ,'DIC' ,'pD' ,'N') return(data.frame(M)) } }
1e992465a200bacb988b9f50678c2e55002a55d3
b068f48befd889ffe4a5e1f7e64ad60177a7348e
/R/legend.R
0fa86e777849daea195466b6c929474a217cccd6
[]
no_license
MaStatLab/BGCR
0495d5e5d260bb5edb31db810fb2371edb32c2c0
7ab3427166f9000cef8604b044aed68676c07bc9
refs/heads/master
2020-06-17T14:24:24.504230
2019-11-24T07:48:47
2019-11-24T07:48:47
195,950,203
3
0
null
null
null
null
UTF-8
R
false
false
1,930
r
legend.R
################################################################################################## ################################################################################################## ####----function to plot the legend of the PMAP plot----#### legend_col = function(col, lev){ opar = par n = length(col) bx = par("usr") box_cx = c(bx[2] + (bx[2] - bx[1]) / 1000, bx[2] + (bx[2] - bx[1]) / 1000 + (bx[2] - bx[1]) / 50) box_sy = (bx[4] - bx[3]) / n box_cy = c(bx[3], bx[3]) xx = rep(box_cx, each = 2) par(xpd = TRUE) for(i in 1:n){ yy = c(box_cy[1] + (box_sy * (i - 1)), box_cy[1] + (box_sy * (i)), box_cy[1] + (box_sy * (i)), box_cy[1] + (box_sy * (i - 1))) polygon(xx, yy, col = col[i], border = col[i]) } par(new = TRUE) plot(0, 0, type = "n", ylim = c(min(lev), max(lev)), yaxt = "n", ylab = "", xaxt = "n", xlab = "", frame.plot = FALSE) axis(side = 4, las = 2, tick =FALSE, line = 0.5, cex.axis=0.7) par = opar } ################################################################################################## ################################################################################################## legend_col_horiz = function(col, lev){ opar = par n = length(col) by = par("usr") box_cy = c(by[3] , by[4]) box_sx = (by[2] - by[1]) / n box_cx = c(by[1], by[1]) yy = rep(box_cy, each = 2) #par(xpd = TRUE) for(i in 1:n){ xx = c(box_cx[1] + (box_sx * (i - 1)), box_cx[1] + (box_sx * (i)), box_cx[1] + (box_sx * (i)), box_cx[1] + (box_sx * (i - 1))) polygon(xx, yy, col = col[i], border = col[i]) } par(new = TRUE) plot(0, 0, type = "n", xlim = c(min(lev), max(lev)), xaxt = "n", ylab = "", yaxt = "n", xlab = "", frame.plot = FALSE) axis(side = 1, las = 1, tick =FALSE, line = 0.5, cex.axis=1) par = opar }
3866d8ee055f7fa78c1586dcff608883f74637c2
4a4d70802bc721f2750e4443723a53c2c37c3225
/gif_animations/emissions_animation/gif_emissions.R
380b0d44dc8c85cbb8344f8d921c2136659fa3c2
[]
no_license
juangordyn/Jamsnot_Vis
4f02b0207e06460c262db689e5caec1d69e3c002
7964e243e935b5cf46f37154d2dac0f3b30453d6
refs/heads/main
2023-08-02T16:39:04.374969
2021-09-30T02:39:21
2021-09-30T02:39:21
411,896,918
0
0
null
null
null
null
UTF-8
R
false
false
1,915
r
gif_emissions.R
library('shiny') library('ggplot2') library('plotly') library('gganimate') library('gifski') library('ggimage') library('dplyr') transport_types <- c('Car', 'Bus', 'Tram', 'Train') text_annotations <- c('184 gCO2/km', '18 gCO2/km ', '13 gCO2/km', '12 gCO2/km') images <- c('car_64.png', 'bus_64.png', 'tram_64.png', 'train_64.png') emissions_df_1 <- data.frame(transport = transport_types, emissions = c(184, 18, 13, 12), frame = rep('b', 4), image = images, label = text_annotations, text_height = c(203, 35, 32, 30 )) emissions_df_2 <- data.frame(transport = transport_types, emissions = c(0, 0, 0, 0), frame = rep('a', 4) , image = images) emissions_df <- bind_rows(emissions_df_2,emissions_df_1) print(emissions_df) emissions_df$transport <- factor(emissions_df$transport, levels = transport_types) key = row.names(emissions_df) key_list = list() for(i in (1:length(key))){ key_list[[i]] <- c(key[i], transport_types) } plot_transport_emission <- ggplot(emissions_df, aes(x = transport, y = emissions, fill = transport, group = 1)) + geom_bar(stat = 'identity', colour = 'black') + scale_fill_manual(values = c("#F64F4C", "orange", "yellow", "#76B756")) + scale_y_continuous(breaks=seq(0,250, 25)) + geom_text(aes(label = label, y = text_height)) + geom_image(aes(image = image), size = 0.075) + theme(plot.title = element_text(face ='bold'), legend.position='none',panel.grid.major = element_blank(), panel.grid.minor = element_blank(), axis.text.x = element_text(face = "bold.italic", color = "black", size = 12)) + labs(x = '' , y='Emissions per km (gCO2/km) per person') + ggtitle('Gas emissions by type of transport') + transition_states(frame, transition_length = 1, state_length = 0, wrap = FALSE) + ease_aes('sine-in-out') anim_save("emissions_animation.gif", animate(plot_transport_emission, duration = 15, end_pause=65))
fd742f638c3109281b000faac61812b1f8dd98a9
a27ddbf3ffafe36488324c1e4a85a542ea526aed
/hw2/jwc/starter.R
5a459a992a1f676a5f5bb80d2eacd1d0cf1697d6
[]
no_license
wclark3/digital_marketing
f7845976f7360ab9ecacee85f318ecfee1d6dffb
a30cda7968646ca3a0f0fbf4ebcdf47ab5f36b13
refs/heads/master
2021-01-17T09:06:38.850769
2016-05-25T04:11:14
2016-05-25T04:11:14
55,690,156
0
1
null
2016-04-26T02:29:32
2016-04-07T11:44:39
R
UTF-8
R
false
false
1,564
r
starter.R
# Digital and Algorithmic Marketing # Homework #2 # Instructor: Sanjog Misra # Topic: Recommendation Systems # if needed: install.packages("recommenderlab) library(recommenderlab) # A Real Dataset data(MovieLense) # Not sure why there is an e at the end MovieLense # Use the Movielens data to answer the following questions # Please upload a pdf file to chalk. Code is not needed. # Hints: To evalaute timing you can use the system.time command. # For example, system.time(r <- Recommender(data, method = "RANDOM") ) # will tell you how gold raining a RANDOM CF algorithm takes. # To assess predictive accuracy you can use the function calcPredictionAccuracy # See ?calcPredictionAccuracy for examples # or the recommenderlab vignette: # https://cran.r-project.org/web/packages/recommenderlab/vignettes/recommenderlab.pdf # Finally, to examine overlap you may use the following function rec.overlap = function(recA,recB,i) { length(intersect(recA@items[[i]],recB@items[[i]])) } # Example # Split the data # Hint: You may want to look at the Reccomderlab # vignette for better ways of doing this. train <- MovieLense[1:300] test <- MovieLense[301:350] # Training the recommender r1 <- Recommender(train, method = "RANDOM") r2 <- Recommender(train, method = "POPULAR") # Predict for the test data rec1 = predict(r1,test,n=150) rec2 = predict(r2,test,n=150) # How many reccomendations overlap for user 12 in the test sample # You can do this for all the users in test rec.overlap(rec1,rec2,i=12)
7f697f49877ee0206b958f2cd529389e703561e9
ce19633f0f8f4dc4fed29d2f211d90f8443d7536
/R_19_01/zad1_19_01.R
ff3a86ff251433717c0d11ef43de38b374b17172
[]
no_license
adrianstodolski/DataScienceCDV
152653e732c432fd58ed42d5888c98acfb0f40bf
4b3e70e59c28bdce153b4dd4aefdc29958c6dcbb
refs/heads/master
2020-09-22T22:40:31.588745
2020-01-25T17:24:05
2020-01-25T17:24:05
225,336,634
0
0
null
null
null
null
UTF-8
R
false
false
1,870
r
zad1_19_01.R
library(ggplot2) # Bibliotek z której będziemy korzystać przy tworzeniu wykresu data_ap <- AirPassengers # plot(data_ap) # widzimy, ze dane maja trend i sezonowosc class(data_ap) str(data_ap) # struktura danych # sprawdzenie danych -> tylko time series start(data_ap) end(data_ap) frequency(data_ap) # naniesienie trendu na wykres plot(data_ap, ylab = "Liczba pasażerów") abline(lm(data_ap ~ time(data_ap))) # analiza danych decompose(data_ap, type = "multiplicative") # rozkladanie danych na skladowe # w rezultacie otrzymamy kilka komponentow # Przedstawiamy dekompozycje na wykresie ap_deco <- decompose(data_ap, type = "multiplicative") plot(ap_deco) ####################### Prognozowanie przyszlosci cycle(data_ap) # dane po miesiacach boxplot(data_ap ~ cycle(data_ap)) # do tej pory dane maja trend plot(data_ap, ylab = "Liczba pasażerów") # dalsz analiza - wyrownujemy dane plot(log(data_ap)) # usunmy trend (funkcja diff) - stosujemy tyle razy az uzyskamy stala srednia plot(diff(log(data_ap))) # Przechodzimy do tworzenia modelu # model ARIMA # Jest to zintegrowania(I) regresji(AR) i sredniej kraczacej(MA) # aby uzyc model mamy 3 parametry # AR -> p # I -> d # MA -> q acf(diff(log(data_ap))) pacf(diff(log(data_ap))) plot(diff(log(data_ap))) # Tworzymy model mod <- arima(log(data_ap), c(0,1,1), seasonal = list(order = c(0,1,1), period = 12)) # Prognozujemy dane na 5 lat mod_pred <- predict(mod, n.ahead = 5*12) summary(mod) pred1 <- 2.718^mod_pred$pred ts.plot(data_ap,pred1,lty=c(1,3)) #lty 1,3 typ lini : 1 oraz 3 ap_szkoleniowe <- ts(data_ap, frequency = 12, start = c(1949,1), end =c(1959,12)) model_1 <- arima(log(ap_szkoleniowe), c(0,1,1), seasonal = list(order=c(0,1,1), period = 12)) ap_pred <- predict(model_1, n.ahead = 1*12) ap_pred <- 2.718^ap_pred$pred ap_pred ap_pred=round(ap_pred,digit=0) #Porównajmy data_ap ap_pred
5276ad24fc4d6100a7256575894dd57256acb9ab
b46a1f0322e9f5ae34fa8e6192b73bf6c0339aef
/man/importance.Rd
3ec05d0cf8ecb42d3574d58944824d5e6cec3c34
[]
no_license
cran/labdsv
33f478ed6469edad170b78577d4aafd04d49b7f7
8a8565f635143f62df5d2bbae3debbd275d5d45e
refs/heads/master
2023-04-14T09:05:46.429809
2023-04-10T08:30:02
2023-04-10T08:30:02
17,696,960
1
0
null
null
null
null
UTF-8
R
false
false
2,174
rd
importance.Rd
\name{importance} \alias{importance} \alias{importance.default} \title{Importance Table} \description{For a classified set of vegetation samples, a importance table lists for each species the average or typical abundance of each species in each class. } \usage{importance(comm,clustering,minval=0,digits=2,show=minval, sort=FALSE,typical=TRUE,spcord,dots=TRUE) } \arguments{ \item{comm}{a data.frame of species abundances with samples as rows and species as columns} \item{clustering}{a vector of (integer) class memberships, or an object of class \sQuote{clustering}, class \sQuote{partana}, of class \code{\link[cluster]{partition}}} \item{minval}{the minimum importance a species must have in at least one class to be included in the output} \item{digits}{the number of digits to report in the table} \item{show}{the minimum value a species must have to print a value} \item{sort}{a switch to control interactive re-ordering} \item{typical}{a switch to control how mean abundance is calculated. Typical=TRUE divides the sum of species abundance by the number of plots in which it occurs; typical=FALSE divides by the number of plots in the type} \item{spcord}{a vector of integers to specify the order in which species should be listed in the table} \item{dots}{a switch to control substituting dots for small values} } \value{a data.frame with species as rows, classes as columns, with average abundance of species in classes.} \note{Importance tables are often used in vegetation classification to calculate or present characteristic species for specific classes or types. Importance may be combined with \code{\link[labdsv]{const}}, \code{\link[labdsv]{concov}} and \code{\link[labdsv]{vegtab}} to achieve a vegetation table-oriented analysis.} \author{ David W. Roberts \email{droberts@montana.edu} } \seealso{\code{\link[labdsv]{const}}, \code{\link[labdsv]{vegtab}}, \code{\link[labdsv]{concov}} } \examples{ data(bryceveg) # returns a data.frame called bryceveg data(brycesite) class <- cut(brycesite$elev,10,labels=FALSE) importance(bryceveg,class,minval=0.25) } \keyword{multivariate}
20f29c54446f1670b593c612a682fd0a14fa9e31
7a281ec4d5cc823b79ebe590a0a2f15be9564df4
/Anime _Shubhasree_Sarkar.R
736fa8b2d5fcb875a5244184e0a9e17ddf4b2dab
[]
no_license
shubhasree1234/Anime-Rating-Prediction
2d3750cec259f7bd4cfc93ae265a2694c8ff432c
330b8de3930a5f60d1a935056cb5d05c820cd71e
refs/heads/main
2023-03-26T14:51:53.062995
2021-03-23T14:52:32
2021-03-23T14:52:32
350,737,415
0
0
null
null
null
null
UTF-8
R
false
false
55,645
r
Anime _Shubhasree_Sarkar.R
#Anime Rating Prediction# #*************************************************************************************************************************# #Step 1-import the libraries library(boot) # library(car) library(QuantPsyc) library(lmtest) library(sandwich) library(vars) library(nortest) library(MASS) library(Hmisc) library(caTools) #used for splitting the data into test and train library(ggplot2) #used for data visualization library(Metrics) library(psych) #*************************************************************************************************************************# #Step 2- Data Extraction or Import #set directory to extract the data setwd("E:\\IVY COURSE\\R STUDIO\\Internship") #use of na.strings in order to convert blank cells to 'NA' Anime<- read.csv(file ="Anime_Final.csv", header = TRUE, fill = T, na.strings = c("","[]")) #get the table containing the dataset View(Anime) #names of the attributes as well as the overall dataset names(Anime) dim(Anime) #check for the dimension of the dataset nrow(Anime) #no of rows ncol(Anime) #no of columns #Result- The dataset is comprised of 7029 rows and 16 columns #*************************************************************************************************************************# #Step 3-check for Data Sanity str(Anime) #check for the overall structure of the data (types of variables) #Result- From the above mentioned code, we could derive that #8 columns are of Character datatype, while rest 7 are integer and one column(Rating) is Numeric summary(Anime) #check for the summarized output in terms of metrics describe(Anime) #check for detailed description of the variables #--------------------------------------------------------------------------------------------------------------------------# #Convert all of the character datatypes to the factor datatypes in the dataset Anime$mediaType<-as.factor(Anime$mediaType) Anime$ongoing<-as.factor(Anime$ongoing) Anime$sznOfRelease<-as.factor(Anime$sznOfRelease) Anime$description<-as.factor(Anime$description) Anime$studios<-as.factor(Anime$studios) Anime$tags<-as.factor(Anime$tags) Anime$contentWarn<-as.factor(Anime$contentWarn) #--------------------------------------------------------------------------------------------------------------------------# #check for data sanity once again str(Anime) #Results - [title(7029 levels),mediaType(9 levels),ongoing(2 levels), # sznOfRelease(5 levels),description(3923 levels), studios(532 levels), # tags(4869 levels),contentWarn(131 levels)] summary(Anime) describe(Anime) #Nature of Variables- #Quantitative -eps,duration,watched,watching,wantWatch,dropped,rating,votes #Categorical- mediaType,ongoing,sznOfRelease, #Qualitative- title,description,studios,tags,contentWarn #*************************************************************************************************************************# #Step 4-Data Pre-processing #the different steps of data preprocessing includes- # 1. Missing Value treatment, 2.Duplicate treatment 3.Outlier treatment #--------------------------------------------------------------------------------------------------------------------------# #4.1-Missing value treatment #check for missing values sapply(Anime, function(x) sum(is.na(x))) # ##Result- From the output we can witness the missing values, # mediaType(30),duration(1696),sznOfRelease(4916) # description(3083), studios(2239),tags(229) # contentWarn(6242), watched(87) #-------------------------------------------------------------------------------------------------------------------------# # check for percentage wise missing values library(dplyr) Missing_values <- data.frame(sapply(Anime, function(y) round((sum(length(which(is.na(y))))/nrow(Anime))*100.00,2))) Missing_values #Filter the missing values >30% Missing_values %>% filter(Missing_values > 30.00) # #Result- Missing values greater than 30% are for the following categories # sznOfRelease- 69.94% # description - 43.86% # studios- 31.85% # contentWarn - 88.80% # for the above cases we will remove them from further calculations #Now for the categorical variables, we will use mode and for numeric variables we will use median #[For the above mentioned columns , as we are going to drop them eventually, hence, it's better # to explore them a bit before dropping ] #--------------------------------------------------------------------------------------------------------------------------# #Exploratory data Analysis #Importing libraries for data visualization and exploration library(tidyverse) library(magrittr) library(sqldf) #--------------------------------------------------------------------------------------------------------------------------# #For the two types of variables , the following graphs or plots are being chosen #Histograms ~ Continuous Variables() #Barplot~ Categorical(sznOfRelease) #Exploration of data using SQL for the Qualitative variables(description,studios,contentWarn) #--------------------------------------------------------------------------------------------------------------------------# #1. Barplot of Season Of Release counts_sznofRelease<- table(Anime$sznOfRelease) barplot(counts_sznofRelease, main = "Barplot of Seasons Of Release" , xlab = "Seasons Of Release" , ylab = "Frequency", col = c("blue","green","yellow","orange","red","lightgreen","gray","violet")) #Results- Spring has the highest no. of Releases followed by Fall #With respect to sznOfRelease, how was Rating distributed boxplot(rating~sznOfRelease, data = Anime, xlab = "Rating", ylab = "sznOfRelease", main = "Rating vs sznOfRelease", horizontal = TRUE) #Results- For the Spring and Summer, there are presence of Outliers, but mostly skewed towards higher rating #--------------------------------------------------------------------------------------------------------------------------# #2. Details on Studios #2.1- Top 5 studios in terms of maximum nos. of Anime viz_1 = data.frame(sqldf('select studios,COUNT(studios) from Anime GROUP BY studios ORDER BY COUNT(studios) desc limit (5)')) viz_1 <- table(viz_1) barplot(viz_1 , main = "Barplot of Studios" , xlab = "Studios" , ylab = "Frequency", col = c("blue","green","yellow","orange","red","lightgreen","gray","violet")) #Results- Sunrise Studios contributed mostly for the Anime Shows #--------------------------------------------------------------------------------------------------------------------------# #3. Details on Top 5 content warnings sqldf('select contentwarn,COUNT(contentwarn) from Anime GROUP BY contentwarn ORDER BY COUNT(contentwarn) desc limit (5)') #Results- The above analysis states Violence being used the most as Contents for shows #--------------------------------------------------------------------------------------------------------------------------# ## Dropping the redundant variables - sznOfRelease,description,studios & contentwarn using subset function Anime<- subset(Anime, select =-c(sznOfRelease,description,studios,contentWarn)) View(Anime) names(Anime) #--------------------------------------------------------------------------------------------------------------------------# #Replacing the missing values #First Step- Categorical Variables ~ replacement using Mode Function #Second Step- Continuous Variables ~ replacement using Median Function #--------------------------------------------------------------------------------------------------------------------------# #First Step-Replacement of categorical variables(tags,mediaType) # Create the function for mode getmode <- function(v) { uniqv <- unique(v) uniqv[which.max(tabulate(match(v, uniqv)))] } # Calculate the mode using the user function. Anime$mediaType[is.na(Anime$mediaType)] <- getmode(Anime$mediaType[!is.na(Anime$mediaType)]) # Mode imputation Anime$tags[is.na(Anime$tags)] <- getmode(Anime$tags[!is.na(Anime$tags)]) # Mode imputation A <- getmode(Anime$mediaType) E <- getmode(Anime$tags) print(A) print(E) View(Anime) #Result- Media type which has 30 missing values, and for tags that has 229 missing values, those are being replaced by the mode function #-------------------------------------------------------------------------------------------------------------------------# #Second Step-Replacement of continuous variables (duration,watched) #replace missing values with median for(i in c(4,11)){ Anime[is.na(Anime[,i]),i]<-round(median(Anime[,i],na.rm = TRUE)) } Anime$watched[is.na(Anime$watched)]<- round(median(Anime$watched,na.rm=TRUE)) #check the output print(str(Anime$watched)) summary(Anime) #Result- Duration which has 1696 missing values, and for watched that has 87 missing values, those are being replaced by the median function #--------------------------------------------------------------------------------------------------------------------------# #4.2-Duplicate values treatment #check for the unique values lapply(Anime, function(x) length(unique(x))) #Result- Tags and Title have too many of the unique values, hence, for further ease in modelling, they should be dropped #Please note that we will backup the data with a different name which we will use to explore the respective attributes present in them #--------------------------------------------------------------------------------------------------------------------------# #back up the data Anime1<-Anime nrow(Anime1) #--------------------------------------------------------------------------------------------------------------------------# #4.3-Outlier treatment #Boxplot view~ Continuous variables) #Rating # identify outliers in r boxplot and save it boxplot(Anime$rating, main = "boxplot for Rating") #quantile view quantile(Anime$rating, seq(0,1, by= 0.005)) #Result- From the Boxplot visualization and also from the quantile processing, it can be confirmed, # that the rating data being continuous, has no outliers present #--------------------------------------------------------------------------------------------------------------------------# #1.Eps- # identify outliers in r boxplot and save it boxplot(Anime$eps, main = "Boxplot of Episodes",xlab = "No. of Episodes", horizontal = TRUE) quantile(Anime$eps, seq(0,1, by= 0.001)) quantile(Anime$eps, c(0.999,0.999301)) #Result- Here, we will be considering the value of 99.9301% which is 399.6171, rounded off to 400, to replace the extreme outliers #check for details length(Anime$eps) summary(Anime$eps) #setting the bench mark eps1 <- as.numeric(round(quantile(Anime$eps,0.999301))) eps1 #Winsorizing Anime$eps = (ifelse(Anime$eps > eps1,eps1,Anime$eps)) #Final visualization summary(Anime$eps) boxplot(Anime$eps,main = "Boxplot of Episodes",xlab = "No. of Episodes", horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #2.Duration- # identify outliers in r boxplot and save it boxplot(Anime$duration,main = "Boxplot of duration",xlab = "Total Duration in minutes", horizontal = TRUE) quantile(Anime$duration, seq(0,1, by= 0.001)) quantile(Anime$duration, c(0.999,0.999701)) #Result- Here, we will be considering the value of 99.9701% which is 150.000, to replace the extreme outliers #check for details length(Anime$duration) summary(Anime$duration) #setting the bench mark dur1 <- as.numeric(round(quantile(Anime$duration,0.999701))) dur1 #Winsorizing Anime$duration= (ifelse(Anime$duration > dur1,dur1,Anime$duration)) #Final visualization summary(Anime$duration) boxplot(Anime$duration,main = "Boxplot of duration",xlab = "Total Duration in minutes", horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #3.Watched- # identify outliers in r boxplot and save it boxplot(Anime$watched, main = "Boxplot of Watched",xlab = "No. of People Watched",horizontal = TRUE) quantile(Anime$watched, seq(0,1, by= 0.001)) quantile(Anime$watched,c(0.999,0.999464)) #Result- Here, we will be considering the value of 99.9464% which is 100019.23 , rounded off to 100019, to replace the extreme outliers(which is 4 in this case) #check for details length(Anime$watched) summary(Anime$watched) #setting the bench mark wat1 <- as.numeric(round(quantile(Anime$watched,0.999464))) wat1 #Winsorizing Anime$watched = (ifelse(Anime$watched > wat1,wat1,Anime$watched)) #Final visualization summary(Anime$watched) boxplot(Anime$watched, main = "Boxplot of Watched",xlab = "No. of People Watched",horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #4.Watching- # identify outliers in r boxplot and save it boxplot(Anime$watching,main = "Boxplot of Watching",xlab = "No. of People Watching", horizontal = TRUE) quantile(Anime$watching, seq(0,1, by= 0.001)) quantile(Anime$watching,c(0.999,0.9995)) #Result- Here, we will be considering the value of 99.95% which is 20021.98 , rounded off to 20022, to replace the extreme outliers(which is 4 in this case) #check for details length(Anime$watching) summary(Anime$watching) #setting the bench mark watchn1 <- as.numeric(round(quantile(Anime$watching,0.9995))) watchn1 #Winsorizing Anime$watching = (ifelse(Anime$watching > watchn1,watchn1,Anime$watching)) #Final visualization summary(Anime$watching) boxplot(Anime$watching,main = "Boxplot of Watching",xlab = "No. of People Watching", horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #5.Wannawatch- # identify outliers in r boxplot and save it boxplot(Anime$wantWatch,main = "Boxplot of Want to Watch",xlab = "No. of People want to watch", horizontal = TRUE) quantile(Anime$wantWatch, seq(0,1, by= 0.001)) quantile(Anime$wantWatch,c(0.999,0.99929)) #Result- Here, we will be considering the value of 99.929% which is 20003.17 , rounded off to 20003, to replace the extreme outliers(which is 5 in this case) #check for details length(Anime$wantWatch) summary(Anime$wantWatch) #setting the bench mark wantw1 <- as.numeric(round(quantile(Anime$wantWatch,0.99929))) wantw1 #Winsorizing Anime$wantWatch = (ifelse(Anime$wantWatch > wantw1,wantw1,Anime$wantWatch)) #Final visualization summary(Anime$wantWatch) boxplot(Anime$wantWatch, main = "Boxplot of Want to Watch",xlab = "No. of People want to watch",horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #6.dropped- # identify outliers in r boxplot and save it boxplot(Anime$dropped, main = "Boxplot of Dropped",xlab = "No. of People dropped out",horizontal = TRUE) quantile(Anime$dropped, seq(0,1, by= 0.001)) quantile(Anime$dropped, c(0.999,0.999381)) #Result- Here, we will be considering the value of 99.9381% which is 4998.508 , rounded off to 4999, to replace the extreme outliers(which is 5 in this case) #check for details length(Anime$dropped) summary(Anime$dropped) #setting the bench mark drp <- as.numeric(round(quantile(Anime$dropped,0.999381))) drp #Winsorizing Anime$dropped = (ifelse(Anime$dropped > drp,drp,Anime$dropped)) #Final visualization summary(Anime$dropped) boxplot(Anime$dropped, main = "Boxplot of Dropped",xlab = "No. of People dropped out", horizontal = TRUE) #-------------------------------------------------------------------------------------------------------------------------# #7.Voted- # identify outliers in r boxplot and save it boxplot(Anime$votes, main = "Boxplot of Votes",xlab = "No. of People Voted", horizontal = TRUE) quantile(Anime$votes, seq(0,1, by= 0.001)) quantile(Anime$votes, c(0.999,0.999454)) #Result- Here, we will be considering the value of 99.9454% which is 80022.01, rounded off to 80022, to replace the extreme outliers(which is 4 in this case) #check for details length(Anime$votes) summary(Anime$votes) #setting the bench mark vote <- as.numeric(round(quantile(Anime$votes,0.999454))) vote #Winsorizing Anime$votes = (ifelse(Anime$votes > vote,vote,Anime$votes)) #Final visualization summary(Anime$votes) boxplot(Anime$votes,main = "Boxplot of Votes",xlab = "No. of People Voted", horizontal = TRUE) #Note- All the outliers are being treated with respect to their quantiles > 99.9% ## Select a subset of Continuous and Categorical variables Cont_Var<- subset(Anime1,select = c(eps,duration,watching,watched,wantWatch,dropped,rating,votes)) Cont_Var Cat_Var<- subset(Anime1,select = c(mediaType,ongoing)) Cat_Var #*************************************************************************************************************************# #Step 5- Exploratory Data Analysis(Graphical) #5.1 Univariate Analysis #*************************************************************************************************************************# #Histograms ~ Continuous Variables() #Barplot~ Categorical(sznOfRelease) #Exploration of data using SQL for the Qualitative variables(description,studios,contentWarn) #*************************************************************************************************************************# ##1. Barplot of Media Types counts_media<- table(Anime1$mediaType) barplot(counts_media, main = "Barplot of MediaTypes" , xlab = "Media types" , ylab = "Frequency", col = c("blue","green","yellow","orange","red","lightgreen","gray","violet")) #Results- TV is the most used medium or platform to watch Anime, while TV special is the least #-------------------------------------------------------------------------------------------------------------------------# ##2. Histogram of Eps hist(Anime1$eps, xlab = "Episodes_distribution", main = "Histogram view of Episodes", col = "blue", xlim = c(0,150), breaks = sqrt(nrow(Anime1))) #Results- Majority of the shows has episodes between 0-20 #-------------------------------------------------------------------------------------------------------------------------# ##3. Histogram of Duration hist(Anime1$duration,xlab = "Duration_distribution",col = "green") hist(Anime1$duration, xlab = "Duration_distribution", main = "Histogram view of Duration", col = "green", xlim = c(0,140), breaks = sqrt(nrow(Anime1))) #Results- Majority of the shows has duration between 0-8 minutes # The Graph over here is Right skewed as observed #-------------------------------------------------------------------------------------------------------------------------# ##4. Barplot of Ongoing counts_ongoing<- table(Anime1$ongoing) barplot(counts_ongoing, main = "Barplot of ongoing shows" , xlab = "Ongoing Shows" , ylab = "Frequency", col = c("blue","green")) #Result- Most of the Shows are not live ones, and hence are not ongoing presently #-------------------------------------------------------------------------------------------------------------------------# ##5. Details on Tags #5.1- Top 5 tags in terms of most nos. of Anime sqldf('select tags,COUNT(tags) from Anime GROUP BY tags ORDER BY COUNT(tags) desc limit (5)') t <- table(Anime1$tags) # frequency of values in tags plot(( sort(t, decreasing=TRUE)[1:5] ), type="h") #Results- Vocaloid is the tag which has being used the most for the Shows #-------------------------------------------------------------------------------------------------------------------------# ##6. Details on Top 5 shows watched by the no. of Users sqldf('select title, SUM(watched) from Anime1 GROUP BY title ORDER BY SUM(watched) desc limit (5)') #Results- Deathnote has been the most watched show ##3. Histogram of watched hist(Anime1$watched,xlab = "watched_distribution", main = "Histogram view of watched",col = "green") hist(Anime1$watched, xlab = "watched_distributionn", main = "Histogram view of watched", col = "green", xlim = c(0,140), breaks = sqrt(nrow(Anime1))) #-------------------------------------------------------------------------------------------------------------------------# ##7.Details on Top 5 shows currently being watched/watching by the no. of Users sqldf('select title, SUM(watching) from Anime1 GROUP BY title ORDER BY SUM(watching) desc limit (5)') #Results- Naruto Shippuden is being the most watched show at present hist(Anime1$watching,xlab = "watching_distribution", main = "Histogram view of watching shows",col = "blue") #-------------------------------------------------------------------------------------------------------------------------# ##8.Details on Top 5 shows people are willing to watch sqldf('select title, SUM(wantWatch) from Anime1 GROUP BY title ORDER BY SUM(wantWatch) desc limit (5)') #Results- Steins;Gate is the show most people are willing to watch hist(Anime1$wantWatch,xlab = "wantwatch_distribution", main = "Histogram view of wantwatch shows",col = "red") #-------------------------------------------------------------------------------------------------------------------------# ##9.Details on Top 5 shows people have dropped before completing it sqldf('select title, SUM(dropped) from Anime1 GROUP BY title ORDER BY SUM(dropped) desc limit (5)') #Results- Naruto Shippuden is the show most people have dropped before completion hist(Anime1$dropped,xlab = "dropped_distribution", main = "Histogram view of dropped shows",col = "red") #-------------------------------------------------------------------------------------------------------------------------# ##10. Histogram of Ratings hist(Anime1$rating, xlab = "Rating", main = "Histogram of Rating", breaks = sqrt(nrow(Anime1)), lines(density(Anime1$rating, adjust=2), lty="dotted", col="darkgreen", lwd=2)) #Result- The rating has Binomial Distribution #-------------------------------------------------------------------------------------------------------------------------# ##11.Details on Top 5 shows which has being voted by users the most sqldf('select title, SUM(votes) from Anime1 GROUP BY title ORDER BY SUM(votes) desc limit (5)') #Results- Death Note is the most voted show by people hist(Anime1$votes,xlab = "votes_distribution", main = "Histogram view of votes of shows",col = "yellow") #*************************************************************************************************************************# #5.2 Bivariate Analysis: #*************************************************************************************************************************# #Scatterplots -Continuous vs Continuous Variables #Boxplots - Continuous vs Categorical Variables #*************************************************************************************************************************# ##1.With respect to mediaType boxplot(rating~mediaType, data = Anime1, xlab = "Rating", ylab = "mediaType", main = "Rating vs mediaType") #Results- For the Movie ,OVA, TV types presence of outliers can be noticed and they are left skewed # More or less, the data has skewness #-------------------------------------------------------------------------------------------------------------------------# ##2.With respect to eps plot(x= Anime1$rating, y= Anime1$eps, xlab = "Rating", ylab = "eps", main = "Rating vs eps") #Results- This graph shows, the shows with episodes between 0-15 have rating between 1.5-3.5 on an average #-------------------------------------------------------------------------------------------------------------------------# ##3.With respect to duration plot(x= Anime1$rating, y= Anime1$duration, xlab = "Rating", ylab = "duration", main = "Rating vs duration") #Results- This graph shows, the shows with less duration have higher ratings , but also many of those with higher durations are rated well. # So no clear interpretation can be done #-------------------------------------------------------------------------------------------------------------------------# ##4.With respect to ongoing boxplot(rating~ongoing, data = Anime1, xlab = "Rating", ylab = "ongoing", main = "Rating vs ongoing") #Results- For the Ongoing shows, presence of outliers can be noticed and they are left skewed # The shows that are not outgoing , have normal distribution #-------------------------------------------------------------------------------------------------------------------------# ##5.With respect to tags plot(x= Anime1$rating, y= Anime1$tags, xlab = "Rating", ylab = "tags", main = "Rating vs tags") #Result- No interpretation can be drawn out of this variable #-------------------------------------------------------------------------------------------------------------------------# ##6.With respect to watched plot(x= Anime1$rating, y= Anime1$watched, xlab = "Rating", ylab = "watched", main = "Rating vs watched") #Result- Most of the high Rated shows are less watched, with few exceptions #-------------------------------------------------------------------------------------------------------------------------# ##7.With respect to watching plot(x= Anime1$rating, y= Anime1$watching, xlab = "Rating", ylab = "watching", main = "Rating vs watching") #Result- Most of the high Rated shows are being currently less watched, with few exceptions #-------------------------------------------------------------------------------------------------------------------------# ##8.With respect to wantWatch plot(x= Anime1$rating, y= Anime1$wantWatch, xlab = "Rating", ylab = "wantWatch", main = "Rating vs wantWatch") #Result- Most of the high Rated shows are in the watchlist of no. of users between 0-10000 #-------------------------------------------------------------------------------------------------------------------------# ##9.With respect to dropped plot(x= Anime1$rating, y= Anime1$dropped, xlab = "Rating", ylab = "dropped", main = "Rating vs dropped") #Result- Most of the high Rated shows are less dropped, w.r.t. others #-------------------------------------------------------------------------------------------------------------------------# ##10.With respect to votes plot(x= Anime1$rating, y= Anime1$votes, xlab = "Rating", ylab = "votes", main = "Rating vs votes") #Result- Most of the high Rated shows are less voted,,there is a gradual trend considering an increase in votes for few of the shows #-------------------------------------------------------------------------------------------------------------------------# ##11.With respect to ongoing boxplot(rating~ongoing, data = Anime1, xlab = "Rating", ylab = "ongoing", main = "Rating vs ongoing") #*************************************************************************************************************************# #Step 6- Exploratory Data Analysis(Statistical tests) #6.1.Correlation tests #Correlation test is used to evaluate the association between two or more variables(continuous) #Pearson correlation(r) -It measures a linear dependence between two variables (x and y). # It's also known as a parametric correlation test because it depends to the distribution of the data. # It can be used only when x and y are from normal distribution. # If the p-value is < 5%, then the correlation between x and y is significant. # Correlation coefficient can be computed using the functions cor() or cor.test() #*************************************************************************************************************************# ##1.With respect to eps Eps_Rate <- cor.test(Anime$rating, Anime$eps, method = "pearson") Eps_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05 # cor of 0.1109282 or 11.09% indicates very weak positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##2.With respect to duration Dur_Rate <- cor.test(Anime$rating, Anime$duration, method = "pearson") Dur_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.2836987 or 28.36% indicates a weak positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##3.With respect to watched Wat_Rate <- cor.test(Anime$rating, Anime$watched, method = "pearson") Wat_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.4267195 or 42.67% indicates a moderate positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##4.With respect to watching Watchn_Rate <- cor.test(Anime$rating, Anime$watching, method = "pearson") Watchn_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.3308784 or 33.08% indicates a weak positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##5.With respect to wantWatch Want_Watch_Rate <- cor.test(Anime$rating, Anime$wantWatch, method = "pearson") Want_Watch_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.5486213 or 54.86% indicates a moderate positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##6.With respect to dropped Drop_Rate <- cor.test(Anime$rating, Anime$dropped, method = "pearson") Drop_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.3286063 or 32.86% indicates a weak positive correlation between the two variables #-------------------------------------------------------------------------------------------------------------------------# ##7.With respect to votes Vote_Rate <- cor.test(Anime$rating, Anime$votes, method = "pearson") Vote_Rate #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means its significant # cor of 0.4154198 or 41.54% indicates a moderate positive correlation between the two variables ## Plot a correlation graph #install.packages("corrplot") library(corrplot) Anime_Cor = cor(Anime[,c("eps","duration","watched","watching","wantWatch","dropped","rating","votes")]) corrplot(Anime_Cor, method = "number") #Considering watched,wantWatch, Votes having correlation >0.40, we need to use log transformation of the same #Consider dropping of Variables with little or no correlation Ex- eps, duration Anime<- subset(Anime, select =-c(eps,duration)) View(Anime) #*************************************************************************************************************************# #6.2.Anova tests #ANOVA stands for Analysis of Variance, is a statistical test used to # analyze the difference between the means of more than two groups(Categorical vs continuous) #1.With respect to MediaTypes,ongoing, tags Anov <- aov(rating ~ mediaType+ongoing+tags, data = Anime) summary(Anov) #Result- P value < 2.2e-16, which is less than the significance level alpha = 0.05, means for both the variables, there is presence of correlation # with respect to the categorical variables as stated # Since the F statistic, in each case is greater than the critical value, # we conclude that there is a significant batch effect at the 0.05 level of significance. #*************************************************************************************************************************# #6.3.Check for Nonlinearity(on the Backed up data) pairs.panels(Cont_Var,col ="red") pairs.panels(Cat_Var,col ="red") View(Anime) #change the values of all numerical variables except the rating using the logarithmic transformation Anime_Cor_1 = cor(Anime[,c("watched","wantWatch","watching", "dropped", "rating","votes")]) corrplot(Anime_Cor_1, method = "number") Anime$watched<-log(Anime$watched) Anime$wantWatch<-log(Anime$wantWatch) Anime$votes<-log(Anime$votes) View(Anime) hist(Anime$wantWatch) # Replace inf with 1 Anime$watched[Anime$watched == -Inf]<- 1 Anime$wantWatch[Anime$wantWatch == -Inf]<- 1 Cont1_Var<- subset(Anime,select = c(eps,duration,watching,watched,wantWatch,dropped,rating,votes)) Cont1_Var pairs.panels(Cont1_Var,col ="red") #*************************************************************************************************************************# # Task 2 #Step 7-Model Development #create a backup data data<- Anime #7.1-Splitting the data set into train and test data set.seed(123) library(caTools) sample<- sample.split(Anime, SplitRatio = 0.70) #-------------------------------------------------------------------------------------------------------------------------# #7.2defining train & test data Anime_train<- subset(Anime, sample == TRUE) Anime_test<- subset(Anime, sample == FALSE) #-------------------------------------------------------------------------------------------------------------------------# #7.3-checking for the no. of rows and columns of the respective datasets nrow(Anime_train) nrow(Anime_test) names(Anime) #-------------------------------------------------------------------------------------------------------------------------# #7.4-create linear regression model Anime_fit<- lm(rating ~., data = Anime_train) ##Step 1 Anime_fit_1 <- lm(rating ~ mediaType + eps + duration + ongoing + tags + watched+ watching+ wantWatch + dropped + votes, data = Anime_train) summary(Anime_fit_1) #-------------------------------------------------------------------------------------------------------------------------# ##Step 2- removing tags Anime_fit_1 <- lm(rating ~ mediaType + ongoing + watched + wantWatch + watching + dropped + votes, data = Anime_train) summary(Anime_fit_1) #-------------------------------------------------------------------------------------------------------------------------# ##Step 3 -removing all other media type except music video,Tv, and Votes Anime_fit_1 <- lm(rating ~ I(mediaType== "Movie") + I(mediaType== "Music Video") + I(mediaType== "TV Special")+ I(ongoing== "Yes") + watched + wantWatch + watching + dropped + votes, data = Anime_train) summary(Anime_fit_1) #-------------------------------------------------------------------------------------------------------------------------# ##Step 4 -removing votes Anime_fit_1 <- lm(rating ~ I(mediaType== "Movie") + I(mediaType== "Music Video") + I(mediaType== "TV Special")+ I(ongoing== "Yes") + watched + wantWatch + watching + dropped, data = Anime_train) summary(Anime_fit_1) #-------------------------------------------------------------------------------------------------------------------------# #check Vif values vif(Anime_fit_1) #Results- Watched and Wantwatch column have high vif, hence should be dropped from the model #-------------------------------------------------------------------------------------------------------------------------# ##Step 5 -removing wantwatch which have high VIF and also correlation is >.40 Anime_fit <- lm(rating ~ I(mediaType== "Movie") + I(ongoing== "Yes") + watched + watching + dropped, data = Anime_train) summary(Anime_fit) #Result- p value of watching is >0.05, hence to be removed crPlots(Anime_fit) #-------------------------------------------------------------------------------------------------------------------------# ##Step 6- removing watching Anime_fit <- lm(rating ~ I(mediaType== "Movie") + I(mediaType== "Music Video") +I(mediaType== "Other") + I(mediaType== "OVA") + I(mediaType== "Web")+ eps + duration + I(ongoing== "Yes") + wantWatch + dropped , data = Anime_train) summary(Anime_fit) abline(Anime_fit,lwd = 3, col = "red") ggplot(Anime_train, aes(rating, dropped) ) + geom_point() + stat_smooth() Anime_fit_2 <- lm(rating ~ I((mediaType== "Movie")^2) + I((mediaType== "Music Video")^2) +I((mediaType== "Other")^2) + I((mediaType== "OVA")^2) + I((mediaType== "Web")^2)+ I(duration^2) + I((ongoing== "Yes")^2) + I(wantWatch^2) + I(dropped^2), data = Anime_train) summary(Anime_fit_2) #*************************************************************************************************************************# #About the results #*************************************************************************************************************************# #The above table proves that there is a weak negative relationship between Rating and mediaType(Movie,Music Video,Other,Web),dropped # and weak positive relationship between Rating and mediaType(OVA),eps,duration,ongoing(Yes),wantWatch #Residual standard error (RSE). #The RSE (or model sigma), corresponding to the prediction error, represents roughly the average difference between the observed outcome values and the predicted values by the model. #The lower the RSE the best the model fits to the data. #Dividing the RSE by the average value of the outcome variable will give us the prediction error rate, which should be as small as possible. #Here the RSE = 0.6784, meaning that the observed rating values deviate from the predicted values by approximately 0.5939 units in average. mean(Anime_train$rating) #3.049757 Error_Rate <-0.5969/mean(Anime_train$rating) Error_Rate #This corresponds to an error rate of 0.6784/mean(Anime_train$rating) = 0.5915/3.049757 = 0.1939499 or 19.39%, which is low. #-------------------------------------------------------------------------------------------------------------------------# #R-squared #The R-squared (R2) ranges from 0 to 1 and represents the proportion of variation in the outcome variable that can be explained by the model predictor variables. #For a simple linear regression, R2 is the square of the Pearson correlation coefficient between # the outcome and the predictor variables. #In multiple linear regression, the R2 represents the correlation coefficient between the observed outcome values and the predicted values. #The R2 measures, how well the model fits the data. The higher the R2, the better the model. #Result- The R square value is 0.5764, which means the moderately okay #-------------------------------------------------------------------------------------------------------------------------# #Adjusted R squared #The adjustment in the Adjusted R Squared value in the summary output is a correction for the number of x variables included in the predictive model. #So the adjusted R-squared is considered, which is a penalized R2 for a higher number of predictors. #An (adjusted) R2 that is close to 1 indicates that a large proportion of the variability in the outcome has been explained by the regression model. #A number near 0 indicates that the regression model did not explain much of the variability in the outcome. #Result- The Adjusted R square value is 0.5756 , which means the moderately okay #-------------------------------------------------------------------------------------------------------------------------# #F-Statistic: #The F-statistic gives the overall significance of the model. #It assess whether at least one predictor variable has a non-zero coefficient. #The F-statistic becomes more important once we start using multiple predictors as in multiple linear regression. #A large F-statistic will corresponds to a statistically significant p-value (p < 0.05). #Results- In this example, the F-statistic equal 706.9 producing a p-value of < 2.2e-16, which is highly significant. #**************************************************************************************************************************# #7.5-estimate the model performance #A more conventional way to estimate the model performance is to display the residual against different measures. par(mfrow=c(2,2)) plot(Anime_fit) #Step 8-check for model validation using correlation plots #Result- It shows many of the variables are highly correlated, which is affecting the model accuracy #8.2-Get the predicted or fitted values fitted(Anime_fit) Anime_train$pred <- fitted(Anime_fit) #Plot the values plot(Anime_train$pred,Anime_train$rating, main= "Actual vs predictive for train data") abline(a=0, b=1) Anime_train$pred_1<- fitted(Anime_fit_2) plot(Anime_train$pred_1,Anime_train$rating, main= "Actual vs predictive for train data") abline(a=0, b=1) #Result- For the rating values between range of 1-4, they are highly concentrated towards the predicted values of 3-4 #**************************************************************************************************************************# #Step 9- Check respective metrics- MAPE, MdAPE, RMSE #9.1-checking MAPE #MAPE- One of the most common metrics used to measure the forecasting accuracy of a model is MAPE, # which stands for mean absolute percentage error. attach(Anime_train) mape(Anime_train$rating,Anime_train$pred) #Results - The MAPE value is 0.1857621 or 18.57% which means the average difference between # the predicted value and the actual value is 18.57%, which is acceptable Mean_Accuracy_train = (1-mape(Anime_train$rating,Anime_train$pred))*100 Mean_Accuracy_train # In other words the mean accuracy thus is (100-MAPE)= (100-18.57)% = 81.42379 #-------------------------------------------------------------------------------------------------------------------------# #9.2-checking Median APE #MdAPE- The Median Absolute Percentage Error (MdAPE) is found by ordering the absolute percentage error (APE) from the smallest to the largest, and using its middle value # (or the average of the middle two values if N is an even number) as the median #MdAPE is more resilient to outliers than MAPE #MdAPE is less intuitive, for example an MdAPE of 8% does not mean that the average absolute percentage error is 8%. # Instead it means that half of the absolute percentage errors are less than 8% and half are over 8% median(abs((Anime_train$rating-Anime_train$pred)/Anime_train$rating)) #Results- The MdAPE value is 0.123073 or 12.30%, which means half of the absolute percentage errors are less than 15.38% # while the rest is over 12.30% Median_Accuracy_train = (1-median(abs((Anime_train$rating-Anime_train$pred)/Anime_train$rating)))*100 Median_Accuracy_train #Thus Median Accuracy is 87.92226 #-------------------------------------------------------------------------------------------------------------------------# #9.3-checking RMSE #Root Mean Square Error-It is the standard deviation of the residuals (prediction errors). # RMSE is a measure of how spread out these residuals are. # It shows how concentrated the data is around the line of best fit. rmse(Anime_train$rating,Anime_train$pred) #Result- RMSE value is0.5908401,thus >0.5 reflects the poor ability of the model to accurately predict the data, which is observed from this dataset #-------------------------------------------------------------------------------------------------------------------------# #9.4-Residuals- A residual is the vertical distance between a data point and the regression line. # Each data point has one residual. # They are positive if they are above the regression line and negative if they are below the regression line. #check for residuals Anime_train$res<-studres(Anime_fit) Anime_train$res install.packages("gvlma") library(gvlma) gvmodel<- gvlma(Anime_fit) summary(gvmodel) install.packages("olsrr") library(olsrr) ols_plot_resid_qq(Anime_fit_2) ols_plot_resid_fit(Anime_fit_2) ols_plot_resid_hist(Anime_fit_2) ols_test_normality(Anime_fit_2) ##################################### Checking of Assumption ############################################ #Step 10- Check for Assumptions # residuals should be uncorrelated ##Autocorrelation # Null H0: residuals from a linear regression are uncorrelated. Value should be close to 2. #Less than 1 and greater than 3 -> concern ## Should get a high p value #10.1-check for multicollinearity #Multicollinearity- Multicollinearity occurs when independent variables in a regression model are correlated. # This correlation is a problem because independent variables should be independent. #VIF- The Variance Inflation Factor (VIF) measures the impact of collinearity among the variables in a regression model. # As per VIF, 1 = not correlated, Between 1 and 5 = moderately correlated, Greater than 5 = highly correlated. vif(Anime_fit) #Conclusion- The vif values being less than 2 for Mediatype,eps,Duration,ongoing, hence we can say, that multicollinearity doesn't exist for them # But for others , like watching, wantwatch, dropped, it is <4, thus acceptable #**************************************************************************************************************************# #10.2-check for autocorrelation #Autocorrelation- Autocorrelation refers to the degree of correlation between # the values of the same variables across different observations in the data. #Ho= the residuals are not correlated, ie. they are independent #H1= the residuals are correlated #Durbin Watson statistic- The Durbin Watson Test is a test statistic used in statistics to detect autocorrelation, # in the residuals from a regression analysis. # The Durbin-Watson statistic will always have a value between 0 and 4. #A value of 2.0 means that there is no autocorrelation detected in the sample. #Values from 0 to less than 2 indicate positive autocorrelation and values from 2 to 4 indicate negative autocorrelation. durbinWatsonTest(Anime_fit) #Conclusion- The D-W Statistic is 1.062099, which is less than 2 and it means that positive autocorrelation exits #**************************************************************************************************************************# #10.3- check for heteroscedasticity #heteroscedasticity- Heteroscedasticity is a systematic change in the spread of the residuals over the # range of measured values and is being caused mainly by the presence of outliers # Breusch-Pagan test - It is used to test for heteroskedasticity in a linear regression model and assumes that the error terms are normally distributed. # It tests whether the variance of the errors from a regression is dependent on the values of the independent variables. # If the test statistic has a p-value below an appropriate threshold (e.g. p < 0.05) then # the null hypothesis of homoskedasticity is rejected and heteroskedasticity assumed. bptest(Anime_fit) #Result - p value from BP test being < 2.2e-16, which is < 0.05 means Heteroscedasticity exists # Null hypothesis -> error is homogenious (p value should be more than 0.05) #Hence null hypothesis can be rejected and therefore heterscedasticity exists #Cook-Weisberg test --- Car package # hypothesis of constant error variance against the alternative that the error variance changes with the level of the response # p value should be more than 0.05 ncvTest(Anime_fit) #Result - p value from NCV test being 0.00084771 thus < 0.05 means Heteroscedasticity exists #**************************************************************************************************************************# #10.4 -Test for normality #Normality- A normality test is used to determine whether sample data has been drawn from a normally distributed population (within some tolerance). #Null hypothesis - The errors should be normally distributed # expected- the p value should be more than 0.05 #Anderson-Darling Test -The Anderson-Darling test is a statistical test of whether a given # sample of data is drawn from a given probability distribution. # The test rejects the hypothesis of normality when the p-value is less than or equal to 0.05. # Failing the normality test allows you to state with 95% confidence the data does not fit the normal distribution. Anime_resids <- Anime_fit$residuals ad.test(Anime_resids) #Conclusion- P value is 8.232e-11, which is <0.05, hence means that with 95% confidence the data does not fit the normal distribution #**************************************************************************************************************************# #Step 11- Check for the Residuals and their Statistics #distribution of residuals #11.1-Histogram hist(Anime_resids) #Result- The residuals show skewness towards left for the residuals #-------------------------------------------------------------------------------------------------------------------------# #11.2-Quantile- Quantile plot~ Q-Q plot, is a graphical tool to help us assess if a set of data plausibly came from some theoretical distribution such as a Normal, exponential or Uniform distribution. # Also, it helps to determine if two data sets come from populations with a common distribution. # In other words,it is a scatterplot created by plotting two sets of quantiles against one another qqnorm(Anime_resids) qqline(Anime_resids) #Result- If both sets of quantiles came from the same distribution, we should see the points forming a line that's roughly straight. # Similarly here as well, the plot is a straight line # As we can see from this plot the errors follow the straight line well so we can say this assumption is met. ####Testing the model on test data ### #Test the model on the remaining dataset. Anime_test$pred<- predict(Anime_fit, Anime_test) attach(Anime_test) plot(pred,rating, main= "Actual vs predictive for test data") abline(a=0, b=1) #Result - Most of variations can be seen between 2-4 values of the variables #-------------------------------------------------------------------------------------------------------------------------# #Check for Mean APE mape(Anime_test$rating,Anime_test$pred) mean(abs((rating-pred)/rating)) #Result- The MAPE value is 18.56% which means # the average difference between the predicted value and the actual value is 22.16%. Mean_Accuracy_test = (1-mape(Anime_test$rating,Anime_test$pred))*100 Mean_Accuracy_test ## The mean accuracy thus is 81.43577 #-------------------------------------------------------------------------------------------------------------------------# #Check for Median APE median(abs((Anime_test$rating-Anime_test$pred)/Anime_test$rating)) #Result- The MdAPE value is 0.1204413 or 12.04%, which means half of the absolute percentage errors are less than 15.38% # while the rest is over 12.04% Median_Accuracy_test = (1-median(abs((Anime_test$rating-Anime_test$pred)/Anime_test$rating)))*100 Median_Accuracy_test #Thus Median Accuracy is 87.95587 #-------------------------------------------------------------------------------------------------------------------------# #Check for RMSE rmse(Anime_test$rating,Anime_test$pred) #Result- RMSE value is 0.5930023,thus =0.5 reflects the poor ability of the model to accurately predict the data, which is observed from this dataset #**************************************************************************************************************************# #check for residuals Anime_test$res<- Anime_test$rating-Anime_test$pred #-------------------------------------------------------------------------------------------------------------------------# #Plot the residuals plot(Anime_test$res) #Results- Residuals are homogeneously scattered forming a band #-------------------------------------------------------------------------------------------------------------------------# hist(Anime_test$res) #Results- The residuals show normal distribution of the data #-------------------------------------------------------------------------------------------------------------------------# qqnorm(Anime_test$res) qqline(Anime_test$res) #Results- If both sets of quantiles came from the same distribution, we should see the points forming a line that's roughly straight. # Similarly here as well, the plot is a straight line #As we can see from this plot the errors follow the straight line well so we can say this assumption is met. #**************************************************************************************************************************# #Conclusion- Thus from the Metrics values like RSE which is considerably low, but Rsquare and adjusted R square value # state that the predictor variables are not highly significant as per the model is concerned and both having value ~ 44% #The assumptions show the dataset being highly skewed, having multiple correlations and collinearity,and deviates from Normal Distribution as per the oevrall frequency being concerned, # even after treatment of outliers, missing values #The MAPE and MdAPE values are approximately 22% and 15% which means the Accuracy being at a considerable range of 75-80% # Business Recommendation #The overall attributes used in the model as we have seen, had weak to moderate correlation with the target variable ie., Rating #Hence, it is difficult to provide business recommendation from that point of View. #But from the overall seasons of Release, as we have observed, that Shows with Ratings Greater than 4, had their Release around Fall and Spring #From the Media Types, Shows with rating >4, are being released on TV the most
b5a8ad56c5489187c49ca7d3677656231ee3c6d4
f29a2a607de04e1c051f36cd964e6896dd4d4876
/capitulo03/scriptR/Problema-10.R
d0a0581b74a3e14183639363b3efb9d8b7f1a83e
[]
no_license
mmejiap/Simulacion-y-modelamiento
6905dd6a24c7fdd112141c424012beca2d6fd0da
d97f0ac5e156bfa4da157894f063a1ffcfd1fdd6
refs/heads/master
2020-07-19T17:52:46.576040
2019-10-09T23:12:06
2019-10-09T23:12:06
206,489,902
0
0
null
null
null
null
UTF-8
R
false
false
227
r
Problema-10.R
rm(list = ls()) k <- 1000 n <- 1:k u <- runif(k,0,1) cov = cumsum(u*exp(u))/n - (cumsum(u)/n )*(cumsum(exp(u))/n) plot(cov,type='l',xlab='n',ylab = 'valor',main='Problema 10') print(paste("Covarianza -> ",cov[length(u)]))
211a954a77439fa2d851d7e4279537fb96d37660
bb35e7e2662ce999c707c176e3784e0ba610c0e1
/man/tile_merge.Rd
0a8616ad5fac4368936b35bdc36858d73d193b43
[]
no_license
kapitzas/WorldClimTiles
a447816c44210c65a7099bff3e992244f92ea149
c84f06070dfeb45e91a35583979c36f0dde9f055
refs/heads/master
2021-07-21T09:48:33.980355
2021-07-07T08:49:37
2021-07-07T08:49:37
176,862,630
4
0
null
null
null
null
UTF-8
R
false
true
1,743
rd
tile_merge.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/tile_merge.R \name{tile_merge} \alias{tile_merge} \title{Merge worldclim data.} \usage{ tile_merge(tiles, fac) } \arguments{ \item{tiles}{List with 0.5 arc min WorldClim tile stacks (or single rasters) (i.e. produced by \code{tile_get}). Names and number of variables need to match between stacks.} \item{fac}{Factor by which resolution is changed, see details.} } \value{ RasterStack or Raster containing merged WorldClim variable(s). } \description{ This function merges tiles downloaded from WorldClim into a single raster. } \details{ \code{raster::merge} is used for merging, at default settings. That means that in case of overlaps the cell values of the object that comes first in the sequence are retained. Raster resolution can be changed to speed up merging of very high resolution data. For example, when data are SRTM tiles, \code{fac = 10} will resample to a resolution of 0.5 arcmin (bilinear), which corresponds to the resolution of WorldClim data. } \examples{ boundary <- getData("GADM", country = "FRA", level = 0) #Downloads France boundaries tilenames <- tile_name(boundary, name = 'worldclim') #Determines names of the worldclim tiles covering France srtmtilenames <- tile_name(boundary, name = 'srtm') #Determines names of the srtm tiles covering France wctiles <- tile_get(tiles = tilenames, name = 'srtm', var = "bio") #downloads WorldClim tiles covering France srtmtiles <- tile_get(tiles = srtmtilenames, name = 'srtm') #downloads SRTM tiles covering France wcmerged <- tile_merge(wctiles) srtmmerged <- tile_merge(srtmtiles) srtmmerged <- tile_merge(srtmtiles, fac = 10) } \author{ Simon Kapitza \email{simon.statecology@gmail.com} }
553f1af82fa2ebe6ef51b223e2132759904ea152
b1217b56f950ba7425c6a406041585fb97547fff
/tests/testthat/test-cached-post.R
c6a0616b89884035634ed421e14817a78e66f6d9
[]
no_license
cran/httpcache
fcc2df0579135531729ccd1a84d1153ea8387dfc
162cc3b635a0627d29e938965cd3c77de8341066
refs/heads/master
2021-05-04T11:23:06.272418
2021-01-10T22:10:02
2021-01-10T22:10:02
52,335,838
0
0
null
null
null
null
UTF-8
R
false
false
2,082
r
test-cached-post.R
context("cachedPOST") public({ clearCache() test_that("Cache gets set on cachedPOST", { expect_length(cacheKeys(), 0) with_fake_http({ expect_POST(a <<- cachedPOST("https://app.crunch.io/api/"), "https://app.crunch.io/api/") expect_POST(b <<- cachedPOST("https://app.crunch.io/api/", body='{"user":"me"}'), 'https://app.crunch.io/api/ {"user":"me"}') }) expect_length(cacheKeys(), 2) expect_true(setequal(cacheKeys(), c("https://app.crunch.io/api/?POST", "https://app.crunch.io/api/?POST&BODY=aec2de8a85873530777f26424e086337"))) expect_identical(a$url, "https://app.crunch.io/api/") expect_identical(content(b), list(user="me")) }) without_internet({ test_that("When the cache is set, can read from it even with no connection", { ## Now read from cache expect_no_request( expect_identical(cachedPOST("https://app.crunch.io/api/"), a) ) expect_no_request( expect_identical(cachedPOST("https://app.crunch.io/api/", body='{"user":"me"}'), b) ) }) test_that("But uncached() prevents reading from the cache", { uncached({ expect_POST(cachedPOST("https://app.crunch.io/api/"), "https://app.crunch.io/api/") expect_POST(cachedPOST("https://app.crunch.io/api/", body='{"user":"me"}'), 'https://app.crunch.io/api/ {"user":"me"}') }) }) test_that("GETs don't read from cachedPOST cache", { expect_GET(uncached(GET("https://app.crunch.io/api/")), "https://app.crunch.io/api/") }) test_that("And POSTs with different payloads don't read the wrong cache", { expect_POST(cachedPOST("https://app.crunch.io/api/", body="wrong"), "https://app.crunch.io/api/ wrong") }) }) })
58d5b258836c57512e706fb12192fc054ead49a7
4fc01b0aa267a461f9a64e74fc67873aab4d3007
/run_analysis.R
a89ee54cb3108536f806385d05841025331c129d
[]
no_license
trponeill/tidy_data
34757b9cc04cf4dfc1f2e45a184245d70a97c1a1
2dae955c1df503197ef076d20f35abd7327adbb4
refs/heads/master
2016-09-06T05:19:52.568762
2014-09-21T15:19:47
2014-09-21T15:19:47
null
0
0
null
null
null
null
UTF-8
R
false
false
2,565
r
run_analysis.R
###run_analysis.R library(plyr) library(dplyr) library(data.table) #Get data into R to see shape and dimensions X_test <- read.table("./UCI HAR Dataset/test/X_test.txt", header = FALSE, sep="") y_test <- read.table("./UCI HAR Dataset/test/y_test.txt", header = FALSE, sep="") X_train <- read.table("./UCI HAR Dataset/train/X_train.txt", header = FALSE, sep="") y_train <- read.table("./UCI HAR Dataset/train/y_train.txt", header = FALSE, sep="") subject_test <- read.table("./UCI HAR Dataset/test/subject_test.txt", header = FALSE, sep="") subject_train <- read.table("./UCI HAR Dataset/train/subject_train.txt", header = FALSE, sep="") features <- read.table("./UCI HAR Dataset/features.txt", header = FALSE, sep="") activity <- read.table("./UCI HAR Dataset/activity_labels.txt", header = FALSE, sep="") #combine the test and train data sets using rbind Xdata<- rbind(X_test, X_train) ydata<- rbind(y_test, y_train) subjectData<- rbind(subject_test, subject_train) #add meaningful column names colnames(Xdata)=(features$V2) #add feature column names to Xdata colnames(activity)=c("activityID","activityType") ##add column names to activity colnames(ydata)=("activityID") #add column names to ydata colnames(subjectData)=("subjectID") #add column names to subject #subset only features with mean/std using dplyr select XdataSubset1 <-select(Xdata, contains("mean")) #subsets columns with 'mean' XdataSubset2 <-select(Xdata, contains("std")) #subset columns with'std' XdataSubset <-cbind(XdataSubset1, XdataSubset2) #combines columns with bind #add activity names to ydata by using plyr 'join' ydata <- arrange(join(ydata, activity), activityID) ydata$activityID <-NULL #drop ID as not needed #combine data sets into one DataSet <- cbind(ydata, subjectData, XdataSubset) #tidy columns using 'names' expand abbrebiations names<-names(DataSet) names<-gsub("mean", "Mean",names) names<-gsub("std", "StdDev",names) names<-gsub("\\()","",names) names<-gsub("^(t)", "Time",names) names<-gsub("tBody", "TimeBody",names) names<-gsub("^(f)", "Freq",names) names<-gsub("BodyBody", "Body",names) setnames(DataSet, names) #apply mean to tidyDataSet variables by activity and subject tidyDataSet <-tbl_df(DataSet) #prepare for dplyr tidyDataSet <- group_by(tidyDataSet, activityType, subjectID) tidyDataSet <-summarise_each(tidyDataSet,funs(mean)) #write tidyDataSet to txt file 'tidy_data write.table(tidyDataSet, file = "./tidy_data.txt", row.name=FALSE)
16b5d36297abef1135ef8c7fc0cd7c58697f1f3a
4d12d0314f6f156bcae0dc82c3c03dcc7cba813b
/plot3.R
086e21cd0e5fdbef0eb16ede260cb218108b62df
[]
no_license
BreizhZut/ExData_Plotting1
8a7d5c4c9c9d9278da6f44aa8b76c56e91ddfddd
1779532e40679379b0d040c009ab60445cec43c0
refs/heads/master
2021-01-17T08:53:50.610284
2016-05-09T06:03:13
2016-05-09T06:03:13
58,339,801
0
0
null
2016-05-09T01:22:56
2016-05-09T01:22:56
null
UTF-8
R
false
false
1,762
r
plot3.R
## We need package lubridate to convert character into time library(lubridate) library(dplyr) ## write the file to read into a variable fileHCG <- "household_power_consumption.txt" ## test for existance if(!file.exists(fileHCG)){ stop(paste("File ",fileHCG,"is not found in working directory")) } ## Let's read the header first headHCG <- read.table(fileHCG,header=TRUE,sep=";",nrows=1) ## From previous test (check README.md for details) we know the ## they are 2880 rows with dates 2007-02-01 and 2007-02-02 (as d/m/yyyy) ## they were written in the file in sequence starting at row 66637 dataHCG <- read.table(fileHCG,header=TRUE,sep=";",skip=66636,nrows=2880,col.names = names(headHCG)) ## combine date and time into a new data.frame of class time datetime <- data.frame(datetime=dmy_hms(paste(dataHCG$Date,dataHCG$Time))) ## Add datetime to the orignal data.frame as the first column dataHCG <- tbl_df(bind_cols(datetime,dataHCG)) ## Clean up temporary variables rm(headHCG) rm(datetime) ## Plot Energy sub metering as a function of time ## Open png device png(filename="plot3.png") ## Make the plot ## Initialise plot with Sub_metering_1 with(dataHCG, plot(datetime,Sub_metering_1,type="l", xlab="",ylab ="Energy sub metering" ) ) ## Add Sub_metering_2 in red with(dataHCG, points(datetime,Sub_metering_2, type="l",col="red" ) ) ## Add Sub_metering_3 in blue with(dataHCG, points(datetime,Sub_metering_3, type="l",col="blue" ) ) ## Add legend ## Since I don't feel like type I use the grep command to generate the labels legend("topright", legend=grep("Sub_metering",names(dataHCG),value = TRUE), col=c("black","red","blue"),lty=c(1,1,1) ) ## Close png device dev.off()
be259481c7f3b7002a9ab0ef5b200638c8a44185
0ea92a0e1eace26312972c5d2be22ae49b59c99c
/R/modeling/json/om_nhd_model_utils.R
0b285c8378da1afeda38343b253f0986c4e3e276
[]
no_license
HARPgroup/vahydro
852425ccf9271ebe2ff95fb6b48d8b7eb23e473f
73ea02207eee96f574f24db95fad6a2f6a496e69
refs/heads/master
2023-09-04T00:11:29.732366
2023-09-01T18:13:46
2023-09-01T18:13:46
147,414,383
0
0
null
2023-09-08T13:52:26
2018-09-04T20:37:13
R
UTF-8
R
false
false
7,302
r
om_nhd_model_utils.R
nhd_next_up <- function (comid, nhd_network) { next_ups <- sqldf( paste( "select * from nhd_network where tonode in ( select fromnode from nhd_network where comid = ", comid, ")" ) ) return(next_ups) } om_handle_wshed_area <- function(wshed_info) { if ("local_area_sqmi" %in% names(wshed_info)) { area_sqmi = wshed_info$local_area_sqmi } else { if ("area_sqmi" %in% names(wshed_info)) { area_sqmi = wshed_info$area_sqmi } else { if ("areasqkm" %in% names(wshed_info)) { area_sqmi = wshed_info$areasqkm * 0.386102 } else { message("Cannot process. Watershed must have either local_area_sqmi or areasqkm") return(FALSE) } } } return(area_sqmi) } om_watershed_container <- function(wshed_info) { if (!("name" %in% names(wshed_info))) { if ("comid" %in% names(wshed_info)) { wshed_info$name = paste0('nhd_', wshed_info$comid) } else { message("Error: watershed info must have 'name' field") return(FALSE) } } wshed_props = list( name=wshed_info$name, object_class = 'ModelObject' ) area_sqmi = om_handle_wshed_area(wshed_info) if ("rchres_id" %in% names(wshed_info)) { rchres_id = wshed_info['rchres_id'] } else { message("Error: You must include 'rchres_id' (ex: RCHRES_R001) to add container. ") return(FALSE) } wshed_props[['drainage_area_sqmi']] = list( name = 'drainage_area_sqmi', object_class = 'Constant', value = area_sqmi ) wshed_props[['run_mode']] = list( name = 'run_mode', object_class = 'Constant', value = run_mode ) wshed_props[['flow_mode']] = list( name = 'flow_mode', object_class = 'Constant', value = flow_mode ) # inflow and unit area wshed_props[['IVOLin']] = list( name = 'IVOLin', object_class = 'ModelLinkage', right_path = paste0('/STATE/', rchres_id, '/HYDR/IVOL'), link_type = 2 ) # this is a fudge, only valid for headwater segments # till we get DSN 10 in place wshed_props[['Runit']] = list( name = 'Runit', object_class = 'Equation', value='IVOLin / drainage_area_sqmi' ) # Get Local & Upstream model inputs wshed_props[['read_from_children']] = list( name='read_from_children', object_class = 'ModelBroadcast', broadcast_type = 'read', broadcast_channel = 'hydroObject', broadcast_hub = 'self', broadcast_params = list( list("Qtrib","Qtrib"), list("trib_area_sqmi","trib_area_sqmi"), list("child_wd_mgd","wd_mgd") ) ) return(wshed_props) } om_facility_model <- function(facility_info) { if (!("name" %in% names(facility_info))) { message("Error: Facility info must have 'name' field") return(FALSE) } facility_props = list( name=facility_info$name, object_class = 'ModelObject' ) facility_props[['send_to_parent']] = list( name='send_to_parent', object_class = 'ModelBroadcast', broadcast_type = 'send', broadcast_channel = 'hydroObject', broadcast_hub = 'parent', broadcast_params = list( list("wd_mgd","wd_mgd"), list("discharge_mgd","ps_nextdown_mgd") ) ) return(facility_props) } om_nestable_watershed <- function(wshed_info) { if (!("name" %in% names(wshed_info))) { if ("comid" %in% names(wshed_info)) { wshed_info$name = paste0('nhd_', wshed_info$comid) } else { message("Error: watershed info must have 'name' field") return(FALSE) } } nested_props = list( name=wshed_info$name, object_class = 'ModelObject' ) area_sqmi = om_handle_wshed_area(wshed_info) nested_props[['local_area_sqmi']] = list( name='local_area_sqmi', object_class = 'Equation', value=area_sqmi ) # Get Upstream model inputs nested_props[['read_from_children']] = list( name='read_from_children', object_class = 'ModelBroadcast', broadcast_type = 'read', broadcast_channel = 'hydroObject', broadcast_hub = 'self', broadcast_params = list( list("Qtrib","Qtrib"), list("trib_area_sqmi","trib_area_sqmi"), list("child_wd_mgd","wd_mgd") ) ) # simulate flows nested_props[['Qlocal']] = list( name='Qlocal', object_class = 'Equation', value=paste('local_area_sqmi * Runit') ) nested_props[['Qin']] = list( name='Qin', object_class = 'Equation', equation=paste('Qlocal + Qtrib') ) nested_props[['Qout']] = list( name='Qout', object_class = 'Equation', equation=paste('Qin * 1.0') ) # calculate secondary properties nested_props[['drainage_area_sqmi']] = list( name='drainage_area_sqmi', object_class = 'Equation', equation=paste('local_area_sqmi + trib_area_sqmi') ) # send to parent object nested_props[['send_to_parent']] = list( name='send_to_parent', object_class = 'ModelBroadcast', broadcast_type = 'send', broadcast_channel = 'hydroObject', broadcast_hub = 'parent', broadcast_params = list( list("Qout","Qtrib"), list("drainage_area_sqmi","trib_area_sqmi") ) ) return(nested_props) } nhd_model_network <- function (wshed_info, nhd_network, json_network) { comid = wshed_info$comid wshed_name = paste0('nhd_', comid) json_network[[wshed_name]] = list( name=wshed_name, object_class = 'ModelObject' ) # base attributes json_network[[wshed_name]][['local_area_sqmi']] = list( name='local_area_sqmi', object_class = 'Equation', value=paste(wshed_info$areasqkm,' * 0.386102') ) # Get Upstream model inputs json_network[[wshed_name]][['read_from_children']] = list( name='read_from_children', object_class = 'ModelBroadcast', broadcast_type = 'read', broadcast_channel = 'hydroObject', broadcast_hub = 'self', broadcast_params = list( list("Qtrib","Qtrib"), list("trib_area_sqmi","trib_area_sqmi") ) ) # simulate flows json_network[[wshed_name]][['Qlocal']] = list( name='Qlocal', object_class = 'Equation', value=paste('local_area_sqmi * Runit') ) json_network[[wshed_name]][['Qin']] = list( name='Qin', object_class = 'Equation', equation=paste('Qlocal + Qtrib') ) json_network[[wshed_name]][['Qout']] = list( name='Qout', object_class = 'Equation', equation=paste('Qin * 1.0') ) # calculate secondary properties json_network[[wshed_name]][['drainage_area_sqmi']] = list( name='drainage_area_sqmi', object_class = 'Equation', equation=paste('local_area_sqmi + trib_area_sqmi') ) # send to parent object json_network[[wshed_name]][['send_to_parent']] = list( name='send_to_parent', object_class = 'ModelBroadcast', broadcast_type = 'send', broadcast_channel = 'hydroObject', broadcast_hub = 'parent', broadcast_params = list( list("Qout","Qtrib"), list("drainage_area_sqmi","trib_area_sqmi") ) ) next_ups <- nhd_next_up(comid, nhd_network) num_tribs = nrow(next_ups) if (num_tribs > 0) { for (n in 1:num_tribs) { trib_info = next_ups[n,] json_network[[wshed_name]] = nhd_model_network(trib_info, nhd_network, json_network[[wshed_name]]) } } return(json_network) }
ef0843618f694157bffbd9a0a60d7c7bfc6020b4
444936c27b6a76d40c100bdfdbcd6917ec136840
/scripts/seeWhatHappens.r
609e428d2134f3343480e37fb4dc26d8ae67901e
[]
no_license
pickledplum/mpg
bb1c09b2393006513d6120a01284ccde059e182e
8b0f376ba7088164bb2273dbc6083dd9659e6979
refs/heads/master
2016-09-10T20:10:54.655702
2014-03-07T23:51:23
2014-03-07T23:51:23
null
0
0
null
null
null
null
UTF-8
R
false
false
11,153
r
seeWhatHappens.r
#' Mining indicators #' library(RSQLite) library(xts) have_GridExtra = FALSE if( "ggplto2" %in% installed.packages() && "gridExtra" %in% installed.packages() ){ library(ggplot2) require(gridExtra) have_GridExtra = TRUE } else{ logger.warn("No ggplot2 or gridExtra packages re installed. No funcy grid plotting...") } source("../dummy/getUniverse.r") source("../dummy/getTSeries.r") source("logger.r") source("advanceMonths.r") source("getCompoundedReturn.r") # colors in the order of lower to higher bins pallet <- c("blue","dark green","yellow","dark orange","red") seeWhatHappens <- function( controlVar, totalR, periods, nbins, factorName ){ # browser() n_periods <- length(periods) valid_months <- as.character(intersect(substr(index(controlVar),1,7), substr(index(totalR),1,7))) oldest_this_year_month <- min(valid_months) tokens <- as.integer(strsplit(oldest_this_year_month, "-")[[1]]) oldest_yy <- tokens[1] oldest_mm <- tokens[2] latest_this_year_month <- max(valid_months) tokens <- as.integer(strsplit(latest_this_year_month, "-")[[1]]) latest_yy <- tokens[1] latest_mm <- tokens[2] logger.debug(paste("Valid months: min,max,# - ", paste(oldest_this_year_month, latest_this_year_month, n_periods),sep=",")) # iterate from the most recent to the oldest for( this_year_month in valid_months ){ logger.info("Processing ", this_year_month, "...") yy <- rep(0,n_periods) mm <- rep(0,n_periods) factor <- as.vector(controlVar[this_year_month]) #logger.debug("factor: ", paste(factor, collapse=",")) names(factor) <- colnames(controlVar) #logger.debug("Names(factor): ", paste(names(factor), collapse=",")) # Get rid of NAs factor <- factor[!is.na(factor)] if( is.empty(factor) ){ logger.warn("No factor data. Skipping...") next } #logger.debug("factor on ", this_year_month, ":", paste(factor,collapse=",")) # Sort in the ascending order sorted_index <- order(factor, decreasing=FALSE) sorted_pbk <- factor[sorted_index] #logger.debug("Sorted index: ", sorted_index) # collect points to the companies data in each bin range <- length(sorted_pbk) if( range < nbins ){ logger.warn("Not enough data. Bin size is ", nbins, " but there is only ", range, " non-null data points. Skipping...") next } percentile_notations <- vector("numeric", nbins) h <- as.integer(range/nbins) bin_ids <- list() sorted_ids <- names(factor)[sorted_index] for(i in 1:nbins){ begin <- (i-1) * h + 1 end <- begin + h - 1 bin_ids[[i]] <- sorted_ids[begin:end] percentile_notations[i] <- as.integer(h*(i-1)/range*100.) #logger.debug("Bin ", percentile_notations[i], " percentile: ", bin_ids[[i]]) } # Will store the returns for the future periods dailyR <- list() # Extract the year and the month of this iteration tokens <- as.integer(strsplit(this_year_month, "-")[[1]]) yy0 <- tokens[1] mm0 <- tokens[2] # Get the price for each company for this month # Keep carrying the company IDs attached to the values # so that we can filter & sort by IDs later. if( is.empty(totalR[ paste(yy0,mm0,sep="-")] )){ logger.warn("No return data on the reference month. Skipping...") next } r0 <- as.vector(totalR[ paste(yy0,mm0,sep="-")][1,sorted_index]) names(r0) <- sorted_ids #logger.debug("Return on (", this_year_month, ")") #print(r0) # Gather incremental returns by months # Example: if periods = c(1,3,6,12) then, # dailyR[[1]] contains returns from the reference month to the next month period. 0,1,2,3 # dailyR[[2]] contains rturns from the first period to the second period. 4,5,6 # dailyR[[3]] contains 7,8,9,10,11,12 are_we_done = FALSE previous_period <- this_year_month for(period in seq(1,n_periods)) { next_period <- advanceMonths(this_year_month, periods[period]-1) logger.debug("Looking into ", next_period) if( !is.empty((totalR[next_period])) ){ dailyR[[period]] <- totalR[paste(previous_period, next_period, sep="/")] #logger.debug("For period from ", previous_period, " to the end of ", next_period, ": ") #print(dailyR[[period]]) } else{ # put dummy in the slot dummy <- t(data.frame(rep(NA, ncol(totalR)))) rownames(dummy) <- paste(next_period, "-01", sep="") colnames(dummy) <- colnames(totalR) dummy <- as.xts(dummy, by=as.Date(index(dummy))) dailyR[[period]] <- dummy # check if there is really no more data beyond that point tokens <- as.integer(strsplit(next_period, "-")[[1]]) next_yy <- tokens[1] next_mm <- tokens[2] if( next_yy > latest_yy ) { # obvious, this is done are_we_done <- TRUE } else if( next_yy == latest_yy ) { if( next_mm > latest_mm ){ # sure, went beyond are_we_done <- TRUE } } else { # No, it's just missing data for this period logger.warn("Skipping this time period in the future due to the lack of data: ", next_period) next } } #browser() previous_period <- next_period } # We need 12 months in the future to do this marching, or we are done. if( are_we_done ){ logger.warn("No more ", periods[n_periods], " months in the future. We are done.") break } # Report how many stocks in each bin num_companies_in_each_bin <- vector("numeric", nbins) for( bin in 1:nbins ){ num_companies_in_each_bin[bin] <- length(bin_ids[[bin]]) } logger.info("Number of stocks in bins in ascending order of percentile: ", paste(num_companies_in_each_bin, collapse=",")) # Compute compounded return upto the period # compoundedR <- list() projectedPrice <- list() for( period in 1:n_periods ){ #ndays <- as.integer(periods[period]*(260/12)) compoundedR_this_bin <- list() for( bin in 1:nbins ){ #logger.debug(as.integer(h*(bin-1)/range*100.), " percentile") r <- rep(NA, length(bin_ids[[bin]])) names(r) <- bin_ids[[bin]] for( i in 1:length(bin_ids[[bin]]) ){ company <- bin_ids[[bin]][i] returns <- as.vector(dailyR[[period]][,company]) #logger.debug("Visiting ", company) ############################################## # ToDO: Revisit by Sachko, 3/7/2014 # # Not quite sure what to do with # "annualization". Here's the code where # total returns are compounded upto this # month period. Do I need to divide # the daily return by 260, or divide the # M-month compounded R by M/12? # # ############################################## r[i] <- getCompoundedReturn( 1., returns ) #/ 12. * periods[period] } if( is.empty(r[!is.na(r)]) ){ logger.warn("No non-NA returns for this bin in period: ", bin, ", ", period) } # Compounded upon the previous time period. if(period>1){ r <- r * compoundedR[[period-1]][[bin]] } compoundedR_this_bin[[bin]] = r #logger.debug("Compounded return for bin=", bin, ": ") #print(r) } compoundedR[[period]] <- compoundedR_this_bin } #print(compoundedR) # average out across the companies within a bin # # Example: The resulting matrix has the dimensions of nbins by nperiods. # 1mon 3mon 6mon 12mon # 0-20 r11 r12 r13 r14 # 21-40 # 41-60 # 61-80 # 81-100 all_returns <- matrix(NA, nrow=nbins, ncol=n_periods) colnames(all_returns) <- periods for( bin in 1:nbins) { for( period in 1:n_periods) { bin_r <- compoundedR[[period]][[bin]] #/ max(periods) * periods[period] all_returns[bin,period] <- mean(bin_r) } } period_mean <- vector("numeric", length=n_periods) rel_returns <- matrix(NA, nrow=nbins, ncol=n_periods) colnames(rel_returns) <- periods for ( period in 1:n_periods ) { period_mean[period] <- mean(all_returns[,period], na.rm=TRUE) rel_returns[,period] <- all_returns[,period] - period_mean[period] } #print(rel_returns) rel_returns <- rel_returns * 100. # to percent if(TRUE) { y <- rel_returns[!is.na(rel_returns)] if( !is.empty(y) ) { percentiles <- seq(0, 100-as.integer(100/nbins), as.integer(100/nbins)) y_min <- -5 #floor(min(y)) y_max <- 5 #max(ceiling(max(y)), y_min) logger.debug("y min,max: ", y_min, ",", y_max) plot(x=1,y=1, type="n", xlim=c(0,max(periods)), ylim=c(y_min,y_max), axes=FALSE, xlab="Months in Future", ylab="Relative Return (%)") axis(side=1, at=c(0,periods) ) axis(side=2, at=seq(y_min,y_max,1) ) abline(0,0, col="black") for(i in 1:nbins){ points(c(0,periods), c(0,rel_returns[i,]), col=pallet[i], pch="*") lines(c(0,periods), c(0,rel_returns[i,]), col=pallet[i], pch="*") } legend("bottom", paste(percentiles, "%"), col=pallet[1:nbins], lty=1:nbins, pch="*", ncol=5) title(main=paste(this_year_month, " - ", factorName, " vs Returns")) # Delay for view Sys.sleep(1) } else{ logger.warn("All Y values are NAs, skipping plotting for ", this_year_month, " ...") } } } }
ef9eb2e05b095544f9eeb483315efa32aa0c80ba
8e96ff3f9d4f3d274fa4908471456e277ee80618
/data/ramldb/Step3_add_sst_data.R
e15b104f73a459f50e22fb4e127bd954536c5a4f
[]
no_license
cfree14/sst_recruitment
1e96d7e9718a93949f17ed0773d0e2702f4dcc4d
134104d76152f1869209e0eb883036274da156a1
refs/heads/master
2020-04-14T08:45:49.817785
2019-01-12T00:15:25
2019-01-12T00:15:25
163,742,595
0
1
null
null
null
null
UTF-8
R
false
false
2,860
r
Step3_add_sst_data.R
# Use PT data # Use data based on STOCKID and not ASSESSID # Setup ################################################################################ # Clear workspace rm(list = ls()) # Packages library(plyr) library(dplyr) library(reshape2) # Directories datadir <- "data/ramldb/data" plotdir <- "data/ramldb/figures" sstdir <- "/Users/cfree/Dropbox/Chris/Rutgers/projects/productivity/data/sst/data/averages" bndrydir <- "/Users/cfree/Dropbox/Chris/Rutgers/projects/productivity/data/stock_boundaries/data" # Read potential RAM data load(file.path(datadir, "ramldb_stock_recruit_data_trimmed.Rdata")) data_orig <- data stocks_orig <- stocks # Read SST data cobe <- read.csv(paste(sstdir, "ramldb_sst_yearly_cobe.csv", sep="/"), as.is=T) # Read stock boundary key key <- read.csv(file.path(bndrydir, "ramldb_v3.8_stock_boundary_table_v2.csv"), as.is=T) # Read centroids key centroids <- read.csv(file.path(bndrydir, "ramldb_v3.8_stock_boundary_centroids_areas_fixed.csv"), as.is=T) # Build assessid, stockid, lme key key1 <- key %>% select(assessid, stockid) %>% left_join(select(centroids, assessid, lme_name), by="assessid") # Build data ################################################################################ # Format SST data sst <- cobe %>% left_join(select(key1, assessid, stockid, lme_name), by="assessid") %>% select(stockid, lme_name, year, sst_c) # Add SST data data1 <- data_orig %>% # Add SST data left_join(sst, by=c("stockid", "year")) %>% select(stockid, lme_name, year, ssb, r, sst_c) %>% filter(!is.na(sst_c)) %>% # Add order/family data left_join(select(stocks_orig, stockid, order, family), by="stockid") %>% # Arrange columns select(stockid, lme_name, order, family, everything()) # Check sample size check <- data1 %>% group_by(stockid) %>% summarize(nyr=n()) %>% filter(nyr>=20) # Final data stocks <- filter(stocks_orig, stockid %in% check$stockid) # Standardize data data <- data1 %>% # Reduce to final stocks filter(stockid %in% check$stockid) %>% # Standardize SSB, R, and SST group_by(stockid) %>% mutate(r_sd=r/sd(r), ssb_sd=ssb/sd(ssb), # r_sd=exp(log(r) - mean(log(r))), # ssb_sd=exp(log(ssb) - mean(log(ssb))), sst_sd=sst_c-mean(sst_c)) # Confirm standardization aggregate(data$r_sd, by=list(data$stockid), sd) # SD=1 aggregate(data$ssb_sd, by=list(data$stockid), sd) # SD=1 aggregate(data$sst_sd, by=list(data$stockid), mean) # Mean=0 # Plot par(mfrow=c(1,2)) plot(r ~ ssb, data, subset=stockid=="CODGB") plot(r_sd ~ ssb_sd, data, subset=stockid=="CODGB") # Complete freeR::complete(data) freeR::complete(stocks) # Build data ################################################################################ # Export save(data, stocks, file=file.path(datadir, "ramldb_stock_recruit_data_trimmed_sst.Rdata"))
6867ca3046c30b41fac458e401d42ebef07d18ac
e4343700d1efe9c5f573be6f26204d4cd5ca17c8
/inst/applications/boardpack2/server.R
b4c3d7017d1e32c51409804d4c69e9460e81a1e1
[]
no_license
jchenpku/Rtraining
bfe2bb6382dda9402e9530eeb1741a5ff046ca26
8e116878aa6ec7ff6ed232a17d87655bb83befa2
refs/heads/master
2021-06-14T09:32:27.721070
2017-04-20T10:32:53
2017-04-20T10:32:53
null
0
0
null
null
null
null
UTF-8
R
false
false
1,792
r
server.R
############# Prep ################ library(shiny) library(knitr) if(!require(devtools)) install.packages("devtools",repos = "https://cran.rstudio.com") if(!require(Rtraining)) devtools::install_github("stephlocke/Rtraining") library("RSQLite") db <- dbConnect(SQLite(), dbname="Boardpack.sqlite") if(!dbExistsTable(db,"Comments")){ emptyComments<-data.frame(Field=NA_character_, Comment=NA_character_, Date=NA_character_) dbWriteTable(db,name = "Comments" ,value=emptyComments) dbSendQuery(db,"DELETE FROM Comments") } shinyServer(function(input, output, session) { ############# Preview Functionality ################ boardpreview<- reactive( readr::read_file( knit2html(input = "MyBoardPack.Rmd",output = "MyBoardPack.html")) ) output$knitDoc <- renderUI({ HTML(boardpreview() ) }) ############# Generate Functionality ################ output$report = downloadHandler( filename <- paste0("MyBoardPack",format(Sys.Date(),"%Y%m%d"),".pdf"), content <- function(file) { out <- rmarkdown::render("MyBoardPack.Rmd", "pdf_document") file.rename(out, file) # move pdf to file for downloading }, contentType = 'application/pdf' ) ############# DB Functionality ################ comments<- reactive({ time<-Sys.time() data <- data.frame( Field= c("TableD1","ChartD1","TableD2","ChartD2"), Comment=c(input$TableD1,input$ChartD1, input$TableD2,input$ChartD2), Date=format(time)) data }) observeEvent(input$save,{ dbWriteTable(db,name = "Comments" ,value=comments(), append=TRUE) }) output$previouscomments<-renderTable({ input$save dbReadTable(db,"Comments") }) })
5c938d65b142e248a7a3492ec6e109535e9516ee
a52bfb54bda7604534487c2744642da30f634b3b
/FPU68882/SRCS/pml-2.03/doc/trlr.r
e692c2e3e9a8668d7c6d1793b2449ab155c09bdf
[]
no_license
Kochise/atari-docs
292ac844dd951065d7d68d91725d013455b8d3c3
51bbffce0b9930c1d349219b9a4c1d60e53f813f
refs/heads/master
2021-01-22T19:49:13.763992
2017-03-29T18:18:13
2017-03-29T18:18:13
85,242,308
1
1
null
null
null
null
UTF-8
R
false
false
376
r
trlr.r
.bp .ce REFERENCES .sp 3 .nf P. A. Fox, A. D. Hall, and N. L. Schryer, The PORT Mathematical Subroutine Library, ACM Transactions on Mathematical Software, Vol 4, No. 2, June 1978, pp 104-126. .sp 3 Brian Ford, Parameterization of the Environment for Transportable Numerical Software, ACM Transactions on Mathematical Software, Vol 4, No. 2, June 1978, pp 100-103. .bp .xp
1eeb02d91740148d1648aa7e93c55430a5eddb5a
72352674e0c2b9a3e1e13e324e484da7e6e23c07
/tests/testthat/test-thumbnail.R
23cf7a637c9c7b3e8497a6ecacc687cc659f785b
[ "Apache-2.0" ]
permissive
karigar/rgee
85ad501dc40362cc4d6494eac0fdfd0f94d5ae87
bea78daa7195e86d2cb6e90dabb5062d4e6e1d76
refs/heads/master
2022-11-12T06:34:17.280941
2020-07-05T04:25:55
2020-07-05T09:05:34
277,981,630
0
0
null
2020-07-08T03:36:34
2020-07-08T03:36:33
null
UTF-8
R
false
false
4,278
r
test-thumbnail.R
context("rgee: ee_as_thumbnail test") # Pre-checking ------------------------------------------------------ # Google credentials were loaded in the system? skip_if_no_credentials <- function(user) { ee_path <- path.expand(sprintf("~/.config/earthengine/%s", user)) credentials <- list.files( path = ee_path, pattern = "@gmail.com|credentials|GCS_AUTH_FILE.json" ) if (length(credentials) != 3) { skip("All google credentials were not found") } } # Neccesary Python packages were loaded? skip_if_no_pypkg <- function() { have_ee <- reticulate::py_module_available("ee") have_numpy <- reticulate::py_module_available("numpy") if (isFALSE(have_ee)) { skip("ee not available for testing") } if (isFALSE(have_numpy)) { skip("numpy not available for testing") } } # Init Earth Engine just if it is necessary init_rgee <- function() { tryCatch( expr = ee$Image()$getInfo(), error = function(e) { ee_Initialize( email = 'data.colec.fbf@gmail.com', drive = TRUE, gcs = TRUE ) } ) } user <- "data.colec.fbf" skip_if_no_credentials(user) skip_if_no_pypkg() init_rgee() # ------------------------------------------------------------------------- library(raster) library(rgee) library(sf) ### 1. Data dem_palette <- c( "#008435", "#1CAC17", "#48D00C", "#B3E34B", "#F4E467", "#F4C84E", "#D59F3C", "#A36D2D", "#C6A889", "#FFFFFF" ) nc <- st_read(system.file("shp/arequipa.shp", package = "rgee")) sheds <- ee$FeatureCollection("USGS/WBD/2017/HUC06")$ filterBounds(ee$Geometry$Rectangle(-127.18, 19.39, -62.75, 51.29))$ map(function(feature) { num <- ee$Number$parse(feature$get("areasqkm")) return(feature$set("areasqkm", num)) }) image <- ee$Image("CGIAR/SRTM90_V4") region <- nc$geometry[[1]] %>% st_bbox() %>% st_as_sfc() %>% st_set_crs(4326) %>% sf_as_ee() %>% ee$FeatureCollection$geometry() # just one band ----------------------------------------------------------- test_that("ee_as_thumbnail full parameters", { arequipa_dem <- ee_as_thumbnail(image = image, region = region, raster = TRUE, vizparams = list(min = 0, max = 5000)) arequipa_dem <- arequipa_dem * 5000 expect_equal(max(getValues(arequipa_dem)), 5000) }) test_that("ee_as_thumbnail min-max", { # JPEG images mysheds <- ee$Feature(sheds$first())$geometry() shed_dem <- ee_as_thumbnail( image = image, region = mysheds$bounds(), vizparams = list( min = 0, max = 500 ) ) expect_equal(max(shed_dem[[1]]), 0.4470588, tolerance = .002) }) # RGB band ----------------------------------------------------------- test_that("ee_as_thumbnail palette, min-max", { # PNG images arequipa_dem <- ee_as_thumbnail( image = image, region = region, vizparams = list(palette = dem_palette, min = 0, max = 5000) ) arequipa_dem <- arequipa_dem * 5000 expect_equal(max(arequipa_dem[[1]]), 5000, tolerance = 1) }) # RGB band ----------------------------------------------------------- test_that("ee_as_thumbnail region", { # PNG images image_clip <- image$clip(region) arequipa_dem <- ee_as_thumbnail( image = image_clip, region = region, raster = TRUE, vizparams = list( palette = dem_palette, min = 0, max = 5000 ) ) arequipa_dem <- arequipa_dem * 5000 expect_equal(mean(arequipa_dem[1:10,1:10,3]), 1638.17, tolerance = 1) }) # error ----------------------------------------------------------- test_that("ee_as_thumbnail error 01", { # PNG images expect_error(ee_as_thumbnail("ee$Image", region)) }) test_that("ee_as_thumbnail error 02", { # PNG images expect_error(ee_as_thumbnail(image, "ee$Geometry")) }) test_that("ee_as_thumbnail error 03", { # PNG images expect_error(ee_as_thumbnail(image, region$centroid()$buffer(100))) }) # large image ----------------------------------------------------------- test_that("ee_as_thumbnail large image", { # PNG images region <- ee$Geometry$Point(-72.403,-16.08)$buffer(100)$bounds() arequipa_dem <- ee_as_thumbnail( image = image, region = region, dimensions = 3000L ) expect_s3_class(arequipa_dem, "stars") })
a8d35e7b3c61cc035b7eaf1b76b0d3b1ab8d4b72
760ec696df1c77a8aa9b82608d6d4515f21fd0e3
/Testing to see if historical change in earnings affects future stock prices (Public).R
b5e3edf94b845b11233c560b1b2785263dc38b16
[]
no_license
gabistreche/Multi_Factor
8e91b6df52c9a791a2adc31574eed1498dca0a87
e385babdae53868b0ad582996ef6cdc03feb7f57
refs/heads/master
2023-03-18T22:13:05.932184
2021-03-08T05:40:54
2021-03-08T05:40:54
null
0
0
null
null
null
null
UTF-8
R
false
false
7,035
r
Testing to see if historical change in earnings affects future stock prices (Public).R
#trying to see if change in earnings trend is predictive of future stock #prices and library(PerformanceAnalytics) library(quantmod) library(tidyverse) library(Quandl) library(ggthemes) library(gridExtra) library(gtable) library(grid) library(TTR) #create custome ggplot theme #from http://joeystanley.com/blog/custom-themes-in-ggplot2 theme_joey <- function () { theme_bw(base_size=12, base_family="Avenir") %+replace% theme( panel.background = element_blank(), plot.background = element_rect(fill="gray96", colour=NA), legend.background = element_rect(fill="transparent", colour=NA), legend.key = element_rect(fill="transparent", colour=NA) ) } #choose start and end dates start <- as.Date("1900-01-01") end <- Sys.Date() #set api key Quandl.api_key("apikeyhere") Yale <- Quandl("YALE/SPCOMP", api_key="apikeyhere", type = "xts") #create 12 month SMA of Earnings as a leading indicator SP_Earnings <- na.omit(Yale$Earnings) SP_Earnings$SMA12 <- SMA(SP_Earnings, n = 12) SP_Earnings$Year.on.Year.Change <- ROC(SP_Earnings[,1], n = 12) colnames(SP_Earnings) <- c("S&P 500 Earnings", "12-Month Moving Average Earnings", "Year on Year Change") #calculate the price return, dividend yield, monthly equivalent, and total return SP_Earnings$Change_in_trend <- Return.calculate(SP_Earnings$`Year on Year Change`) SP500 <- merge(Yale$`S&P Composite`, Yale$Dividend) SP500$Price_Return <- Return.calculate(SP500$S.P.Composite) SP500$Dividend_Yield <- SP500$Dividend/SP500$S.P.Composite SP500$Dividend_Yield_Monthly <- SP500$Dividend_Yield/12 SP500$Total_Return <- SP500$Price_Return + SP500$Dividend_Yield_Monthly SP500 <- na.omit(SP500) #check to make sure it looks correct head(SP500) table.AnnualizedReturns(SP500[,c(3,5,6)]) #create a date time index that will work with other functions in #performance analytics package and match length of current #data from Yale Shiller. I chose 28th of the month since it #will be at the end of the month and don't have to worry about February #dates <- seq(as.Date("1871-02-28"),length=1778,by="months") #SP500_xts <- cbind(SP500$Total_Return, dates) #SP500_data <- as.xts(SP500_xts$Total_Return, order.by = dates) #colnames(SP500_data) <- "SP500 TR" #SP500_data$rolling_12month_returns <- rollapply(SP500_data[,1], #FUN = Return.annualized, #width = 12, #scale = 12) #SP500_data$rolling_24month_returns <- rollapply(SP500_data[,1], #FUN = Return.annualized, #width = 24, #scale = 12) #SP500_data$rolling_36month_returns <- rollapply(SP500_data[,1], #FUN = Return.annualized, #width = 36, #scale = 12) #check to make sure the data looks correct #autoplot(SP500_data[,c(2,3,4)], facets = FALSE) #lag the total return SP500 performance to correlate with the current month #earnings and the next 12, 24, 36 month returns #SP500_data$rolling_12month_returns_lag <- lag.xts(SP500_data$rolling_12month_returns, k = -12) #SP500_data$rolling_24month_returns_lag <- lag.xts(SP500_data$rolling_24month_returns, k = -24) #SP500_data$rolling_36month_returns_lag <- lag.xts(SP500_data$rolling_36month_returns, k = -36) head(SP500) SP500$rolling_12month_returns <- rollapply(SP500$Total_Return, FUN = Return.annualized, width = 12, scale = 12) SP500$rolling_24month_returns <- rollapply(SP500$Total_Return, FUN = Return.annualized, width = 24, scale = 12) SP500$rolling_36month_returns <- rollapply(SP500$Total_Return, FUN = Return.annualized, width = 36, scale = 12) #lag the total return SP500 performance to correlate with the current month #earnings and the next 12, 24, 36 month returns SP500$rolling_12month_returns_lag <- lag.xts(SP500$rolling_12month_returns, k = -12) SP500$rolling_24month_returns_lag <- lag.xts(SP500$rolling_24month_returns, k = -24) SP500$rolling_36month_returns_lag <- lag.xts(SP500$rolling_36month_returns, k = -36) tail(SP500) Test_data <- merge(SP500, SP_Earnings) Test_data_na <- na.omit(Test_data) tail(Test_data_na, 5) #the chart of correlation seems to indicate that change in earnings on a calendar basis #is not correlated with future returns chart.Correlation(Test_data_na[,c(10,11,12,15,16)]) #create a trading strategy to see if getting in and out depending on earnings #can work #1st signal and strategy is if the change in trend is negative get out signal <- ifelse(Test_data_na$Change_in_trend > 0, 1, 0) #add the signal Test_data_na$signal <- signal #calculate the returns based on the price data times the signals. must add to signal three months to buy on the day #after the signal Test_data_na$signal_lagged <- lag(signal,3) Test_data_na$portfolio.return_no_costs_1 <- (Test_data_na$Total_Return*Test_data_na$signal_lagged) tail(Test_data_na, 30) ncol(Test_data_na) table.AnnualizedReturns(Test_data_na[,c(6,19)]) #second signal is to see if change in trend is greater than -.05 otherwise get out signal2 <- ifelse(Test_data_na$Change_in_trend > -.05, 1, 0) #add the signal Test_data_na$signal2 <- signal2 #calculate the returns based on the price data times the signal. must add to signal three months to buy on the day #after the signal Test_data_na$signal_lagged2 <- lag(signal2,3) Test_data_na$portfolio.return_no_costs_2 <- (Test_data_na$Total_Return*Test_data_na$signal_lagged2) tail(Test_data_na, 30) table.AnnualizedReturns(Test_data_na[,c(6,22)]) #Third signal to see if simple year on year change is greater than -.05 otherwise get out signal3 <- ifelse(Test_data_na$Year.on.Year.Change > -.05, 1, 0) #add the signal Test_data_na$signal3 <- signal3 #calculate the returns based on the price data times the signal. must add to signal three months to buy on the day #after the signal Test_data_na$signal_lagged3 <- lag(signal3,3) Test_data_na$portfolio.return_no_costs_3 <- (Test_data_na$Total_Return*Test_data_na$signal_lagged3) tail(Test_data_na, 30) table.AnnualizedReturns(Test_data_na[,c(6,25)]) chart.CumReturns(Test_data_na[,c(6,19,22,25)], legend.loc = "topleft") #based on a few different signals, even ignoring transaction costs, one #cannot use earnings, change in earnings, or change in earnings trend #to time the market and get superior risk adjusted performance
d54befc0de8be0f7cc64b85bbfbe5f18fb0ffd05
11c18a4332b419d124e331ef1ba0de2191cdb794
/man/gg_colors.Rd
78b5d3354d8c1d40ee8a425db7f11db58ec62423
[]
no_license
MarkKurzeja/rankr
fca771cd35a073e676ba592cd36f9af4017d44fd
1f0b2fe36ea33d73fab08c25f443ab611ea33112
refs/heads/master
2020-12-31T04:42:31.731017
2016-04-28T20:40:37
2016-04-28T20:40:37
57,325,361
0
0
null
null
null
null
UTF-8
R
false
true
434
rd
gg_colors.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/rankr_main.R \name{gg_colors} \alias{gg_colors} \title{ggplot2 Color Defaults} \usage{ gg_colors(n) } \arguments{ \item{n}{number of colors} } \description{ The colors from the great ggplot2 package by Hadley Wickhim which are evenly spaced hues aroudn the hcl colour wheel } \examples{ scales::show_col(gg_colors(4)) } \keyword{colors} \keyword{ggplot2}
8e9a8ab0c13bddfdf618cbf02a7eb0bbaddc39d9
17bd236260351ebc9cf1bff367c0e63dfb3961d0
/man/Plot.CA.Rd
e05aeba42e2059b1b73f1415287ef55afe35574b
[]
no_license
LeoDonatoNunes/MVar.pt
7b68119d79c052ba57025b22e51026cce68ae657
1a95f718564b438bfd65f43b06f977bdde61d775
refs/heads/master
2021-01-17T13:04:15.307181
2015-10-06T22:14:02
2015-10-06T22:14:02
null
0
0
null
null
null
null
UTF-8
R
false
false
1,319
rd
Plot.CA.Rd
\name{Plot.CA} \alias{Plot.CA} \title{Graficos da analise de correspondencia (CA) simples e multipla} \description{Graficos da analise de correspondencia (CA) simples e multipla.} \usage{Plot.CA(AC, Titles = matrix(NA,1,3), Color = "s", LinLab = NULL)} \arguments{ \item{AC}{Dados da funcao CA.} \item{Titles}{Titulos para os graficos.} \item{Color}{"s" para graficos coloridos - default,\cr "n" para graficos em preto e branco.} \item{LinLab}{Vetor com o rotulo para as linhas, para dados de frequencia.\cr Se nao informado retorna o padrao dos dados.} } \value{Retorna varios graficos.} \author{ Paulo Cesar Ossani Marcelo Angelo Cirillo } \seealso{\code{\link{CA}}} \examples{ data(DataFreq) # conjunto de dados de frequencia Dat <- DataFreq[,2:ncol(DataFreq)] rownames(Dat) <- as.character(t(DataFreq[1:nrow(DataFreq),1])) Resp <- CA(Dat, "f") # realiza CA Titles = c("Observacoes", "Variaveis", "Observacoes/Variaveis") Plot.CA(Resp, Titles, Color = "s") data(DataQuali) # Conjunto de dados qualitativos Dat <- DataQuali[,2:ncol(DataQuali)] Resp <- CA(Dat, "c", "b") # realiza CA Titles = c("","Graficos das Variaveis") Plot.CA(Resp, Titles, Color = "s") } \keyword{Analise de correspondencia} \keyword{CA}
7e507b990844b86a224a22c3a072ef8ec27bc3ed
c1d359cdf0281885744cdcd85d41a21e91218b43
/man/sdf_read_table.Rd
e09819998016df15448f43a11f4c6653d0ce8c37
[ "MIT" ]
permissive
zwdzwd/sesame
20b2d29578661487db53432c8991d3c4478aa2c1
62fe6ef99a02e7f94b121fb601c3f368b8a4c1a8
refs/heads/master
2023-08-08T01:45:02.112492
2023-07-26T13:23:03
2023-07-26T13:23:03
122,086,019
37
26
MIT
2023-01-05T16:02:38
2018-02-19T16:00:34
R
UTF-8
R
false
true
700
rd
sdf_read_table.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/SigDFMethods.R \name{sdf_read_table} \alias{sdf_read_table} \title{read a table file to SigDF} \usage{ sdf_read_table(fname, platform = NULL, verbose = FALSE, ...) } \arguments{ \item{fname}{file name} \item{platform}{array platform (will infer if not given)} \item{verbose}{print more information} \item{...}{additional argument to read.table} } \value{ read table file to SigDF } \description{ read a table file to SigDF } \examples{ sesameDataCache() # if not done yet sdf <- sesameDataGet('EPIC.1.SigDF') fname <- sprintf("\%s/sigdf.txt", tempdir()) sdf_write_table(sdf, file=fname) sdf2 <- sdf_read_table(fname) }
1a42b19b350ad802331fe27550c4dc8374fbf751
7f27db9c3a8e1eeda456dc64f11338466c6a2a98
/man/set_methods.Rd
37503f7fbbd581d0ec5b4ef7a0ce417aa6ac5da0
[ "MIT" ]
permissive
ropensci/EML
4c228654f2fbcc5846121255dbb3dc19ba1c61df
b7871cca2b996a33aa1f534e8446f0730e706d4d
refs/heads/master
2023-05-24T01:50:33.364460
2022-06-06T22:10:05
2022-06-06T22:10:05
10,894,022
75
43
NOASSERTION
2022-06-06T22:10:07
2013-06-23T23:20:03
R
UTF-8
R
false
true
1,338
rd
set_methods.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/set_methods.R \name{set_methods} \alias{set_methods} \title{set_methods} \usage{ set_methods( methods_file, instrumentation = character(), software = NULL, sampling_file = NULL, sampling_coverage = NULL, sampling_citation = NULL, qualityControl_file = NULL ) } \arguments{ \item{methods_file}{Path to a file (markdown or .docx) containing a description of the methods used} \item{instrumentation}{optional, text describing instrumentation used in methods} \item{software}{optional, an EML software node describing software used in methods} \item{sampling_file}{optional, Path to a file (.md or .docx) describing sampling method} \item{sampling_coverage}{optional, coverage node for methods, e.g. set_coverage()} \item{sampling_citation}{optional, a citation element describing the sampling protocol} \item{qualityControl_file}{optional, path to a file (.md or .docx) describing quality control methods} } \value{ A methods object } \description{ set_methods } \examples{ \donttest{ f <- system.file("examples/hf205-methods.md", package = "EML") set_methods(methods_file = f) ## Can also import from methods written in a .docx MS Word file. f <- system.file("examples/hf205-methods.docx", package = "EML") set_methods(methods_file = f) } }
21d5d342a987c56714ed355953ff2156d05600d0
323611254fe6aac77f4c519b0c9c757ddb66fb7d
/univarb.R
96315694428a0b30b57bd5986b8bb05ff125a177
[]
no_license
MattHealey/Sudi
976f25e491eafed459e3252391900ec867b6018b
cd280fd3c8060aee150c7d13cbdaf9b6e953e246
refs/heads/master
2021-01-16T21:37:06.830794
2016-07-10T12:02:17
2016-07-10T12:02:17
62,262,532
0
0
null
null
null
null
UTF-8
R
false
false
19,254
r
univarb.R
options(scipen = 999) #source("1 - clean.R") library(multcomp) library(effects) library(lubridate) library(margins) #row.has.na <- apply(usudi, 1, function(x){any(is.na(x))}); sum(row.has.na);magesudi <- usudi[!row.has.na,]; rm(row.has.na) usudi$bwr[usudi$bwr==0] <- NA contrasts(usudi$yod) <- contr.treatment(levels(usudi$yod), base=which(levels(usudi$yod) == "2014")) contrasts(usudi$dhb) <- contr.treatment(levels(usudi$dhb), base=which(levels(usudi$dhb) == "Southern")) contrasts(usudi$eth) <- contr.treatment(levels(usudi$eth), base=which(levels(usudi$eth) == "European_or_other")) contrasts(usudi$sex) <- contr.treatment(levels(usudi$sex), base=which(levels(usudi$sex) == "F")) contrasts(usudi$dep) <- contr.treatment(levels(usudi$dep), base=which(levels(usudi$dep) == "01_08")) ## Maternal age ## create factor for mage age <- function(dob, age.day, units = "years", floor = TRUE) { calc.age = interval(dob, age.day) / duration(num = 1, units = units) if (floor) return(as.integer(floor(calc.age))) return(calc.age) } usudi$mage <- age(dob = usudi$mdob, age.day = usudi$dob) #magesudi$mage <- age(dob = magesudi$mdob, age.day = magesudi$dob) addmargins(table(usudi$mage, usudi$sudi, exclude = NULL)) # 0 1 <NA> Sum #11 1 0 0 1 #12 5 0 0 5 #13 51 1 0 52 #14 346 1 0 347 #15 1469 4 0 1473 #16 4727 10 0 4737 #17 10123 16 0 10139 #18 15782 28 0 15810 #19 21320 50 0 21370 #20 24198 21 0 24219 #21 25881 48 0 25929 #22 28255 48 0 28303 #23 30101 42 0 30143 #24 31627 34 0 31661 #25 33389 34 0 33423 #26 35970 27 0 35997 #27 38673 28 0 38701 #28 42016 26 0 42042 #29 44883 13 0 44896 #30 46152 16 0 46168 #31 48188 19 0 48207 #32 47798 11 0 47809 #33 45252 17 0 45269 #34 42327 20 0 42347 #35 38152 16 0 38168 #36 32664 16 0 32680 #37 27024 8 0 27032 #38 21543 5 0 21548 #39 16741 8 0 16749 #40 11980 4 0 11984 #41 7994 0 0 7994 #42 5029 0 0 5029 #43 2885 2 0 2887 #44 1544 1 0 1545 #45 781 1 0 782 #46 360 0 0 360 #47 157 0 0 157 #48 86 0 0 86 #49 30 0 0 30 #50 18 0 0 18 #51 17 0 0 17 #52 5 0 0 5 #53 4 0 0 4 #54 1 0 0 1 #55 3 0 0 3 #56 2 0 0 2 #57 1 0 0 1 #64 1 0 0 1 #65 2 0 0 2 #70 1 0 0 1 #<NA> 8 158 0 166 #Sum 785567 733 0 786300 age.cat <- function(x, lower = 0, upper, by = 5, sep = "-", above.char = "+") { labs <- c(paste(seq(lower, upper - by, by = by), seq(lower + by - 1, upper - 1, by = by), sep = sep), paste(upper, above.char, sep = "")) cut(floor(x), breaks = c(seq(lower, upper, by = by), Inf), right = FALSE, labels = labs) } addmargins(table(age.cat(usudi$mage, lower = 10,upper = 80), usudi$sudi, exclude = NULL)) # 0 1 <NA> Sum #10-14 403 2 0 405 #15-19 53421 108 0 53529 #20-24 140062 193 0 140255 #25-29 194931 128 0 195059 #30-34 229717 83 0 229800 #35-39 136124 53 0 136177 #40-44 29432 7 0 29439 #45-49 1414 1 0 1415 #50-54 45 0 0 45 #55-59 6 0 0 6 #60-64 1 0 0 1 #65-69 2 0 0 2 #70-74 1 0 0 1 #75-79 0 0 0 0 #80+ 0 0 0 0 #<NA> 8 158 0 166 #Sum 785567 733 0 786300 usudi$magecat <- age.cat(usudi$mage, lower = 10,upper = 80) levels(usudi$magecat) #[1] "10-14" "15-19" "20-24" "25-29" "30-34" "35-39" "40-44" "45-49" "50-54" "55-59" "60-64" "65-69" "70-74" "75-79" "80+" levels(usudi$magecat) <- list("under20" = c("10-14","15-19"), "20-24" = "20-24", "25-29" = "25-29", "30-34" = "30-34", "35-39" = "35-39", "over40" = c("40-44","45-49","50-54","55-59","60-64","65-69","70-74","75-79","80+")) sudi <- droplevels(usudi) levels(usudi$magecat) addmargins(table(usudi$magecat, usudi$sudi, exclude = NULL)) # 0 1 <NA> Sum #>20 53824 110 0 53934 #20-24 140062 193 0 140255 #25-29 194931 128 0 195059 #30-34 229717 83 0 229800 #35-39 136124 53 0 136177 #<40 30901 8 0 30909 #<NA> 8 158 0 166 #Sum 785567 733 0 786300 contrasts(usudi$magecat) <- contr.treatment(levels(usudi$magecat), base=which(levels(usudi$magecat) == "over40")) rm(age,age.cat) s1 <- glm(sudi ~ yod, data = usudi, family = binomial(link = "log")) s2 <- glm(sudi ~ bw, data = usudi, family = binomial(link = "log")) s3 <- glm(sudi ~ eth, data = usudi, family = binomial(link = "log")) s4 <- glm(sudi ~ sex, data = usudi, family = binomial(link = "log")) s5 <- glm(sudi ~ dhb, data = usudi, family = binomial(link = "log")) s6 <- glm(sudi ~ dep, data = usudi, family = binomial(link = "log")) s7 <- glm(sudi ~ magecat, data = usudi, family = binomial(link = "log")) s8 <- glm(sudi ~ bwr, data = usudi, family = binomial(link = "log")) s9 <- glm(sudi ~ mage, data = usudi, family = binomial(link = "log")) library(doBy) orderBy(~ AIC, AIC(s1,s2,s3,s4,s5,s6,s7)) ### ## Year of Death # addmargins(table(usudi$yod, usudi$sudi, exclude = NULL)) summary(s1 <- glm(sudi ~ yod, data = usudi, family = binomial(link = "log"))) s1b <- glm(sudi ~ 0 + yod, data = usudi, family = binomial(link = "log")) anova(s1, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 786299 11695 #yod 12 20.823 786287 11674 0.05304 . year.ht <- glht(s1, linfct = mcp(yod = "Tukey")) summary(year.ht) plot(year.ht) plot(allEffects(s1)) plot(margins(s1b, type = "link")[[1]]) plot(margins(s1, type = "link")[[1]]) m <- s1 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) #Est LL UL #(Intercept) 0.0006884563 0.0005050495 0.0009384666 #yod2002 1.4519065953 0.9648031975 2.1849355049 #yod2003 1.6042566267 1.0782473193 2.3868729171 #yod2004 1.5407978031 1.0355920731 2.2924643127 #yod2005 1.2651144223 0.8364064852 1.9135606069 #yod2006 1.7167325587 1.1654455220 2.5287931717 #yod2007 1.4548079293 0.9813302382 2.1567317799 #yod2008 1.4497370441 0.9779094532 2.1492148277 #yod2009 1.4499473158 0.9757329670 2.1546337881 #yod2010 1.3965287039 0.9386195084 2.0778306900 #yod2011 1.2659600051 0.8412309169 1.9051305681 #yod2012 0.9392033882 0.6060188657 1.4555702047 #yod2013 1.0737210826 0.6998363295 1.6473522659 ### ## Birth Weight # addmargins(table(usudi$bw, usudi$sudi, exclude = NULL)) foo <- usudi[usudi$bw != "Unknown",]; foo <- droplevels(foo) contrasts(foo$bw) <- contr.treatment(levels(foo$bw), base=which(levels(foo$bw) == "over4500")) summary(s2 <- glm(sudi ~ bw, data = foo, family = binomial(link = "log"))) s2b <- glm(sudi ~ 0 + bw, data = foo, family = binomial(link = "log")) anova(s2, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 785565 11596 #bw 8 279.75 785557 11316 < 0.00000000000000022 *** bw.ht <- glht(s2, linfct = mcp(bw = "Tukey")) summary(bw.ht) plot(bw.ht) plot(allEffects(s2)) plot(margins(s2b, type = "link")[[1]]) plot(margins(s2, type = "link")[[1]]) m <- s2 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.000095225 0.0000238706 0.0003798732 #bwunder1000 16.842732491 3.4073282582 83.2551536765 #bw1000-1499 31.475104241 7.1709458784 138.1522331592 #bw1500-1999 35.643423701 8.5624717940 148.3746380380 #bw2000-2499 28.418252763 7.0007755510 115.3582319861 #bw2500-2999 17.313116111 4.3058396852 69.6133649608 #bw3000-3499 9.880542360 2.4628593436 39.6389333340 #bw3500-3999 5.525348281 1.3707497435 22.2720987346 #bw4000-4499 4.277268915 1.0360875570 17.6578024183 ### ## Ethnicity # addmargins(table(usudi$eth, usudi$sudi, exclude = NULL)) foo <- usudi[usudi$eth != "Unknown",]; foo <- droplevels(foo) contrasts(foo$eth) <- contr.treatment(levels(foo$eth), base=which(levels(foo$eth) == "European_or_other")) summary(s3 <- glm(sudi ~ eth, data = foo, family = binomial(link = "log"))) s3b <- glm(sudi ~ 0 + eth, data = foo, family = binomial(link = "log")) anova(s3, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 785716 11680 #eth 2 541.64 785714 11138 < 0.00000000000000022 *** eth.ht <- glht(s3, linfct = mcp(eth = "Tukey")) summary(eth.ht) plot(eth.ht) plot(allEffects(s3)) plot(margins(s3b)[[1]]) plot(margins(s3)[[1]]) m <- s3 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.0003089326 0.0002626802 0.000363329 #ethMaori 7.0387997870 5.8527777958 8.465160334 #ethPacific 3.4457815638 2.6524246786 4.476436478 ### ## Sex # addmargins(table(usudi$sex, usudi$sudi, exclude = NULL)) summary(s4 <- glm(sudi ~ sex, data = usudi, family = binomial(link = "log"))) s4b <- glm(sudi ~ 0 + sex, data = usudi, family = binomial(link = "log")) anova(s4, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 786299 11695 #sex 1 11.028 786298 11684 0.0008975 *** sex.ht <- glht(s4, linfct = mcp(sex = "Tukey")) summary(sex.ht) plot(sex.ht) plot(allEffects(s4)) plot(margins(s4b)[[1]]) m <- s4 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.0008151365 0.0007295571 0.0009107545 #sexM 1.2798616806 1.1056158159 1.4815688216 ### ## Location # addmargins(table(usudi$dhb, usudi$sudi, exclude = NULL)) foo <- usudi[usudi$dhb != "Unknown",]; foo <- droplevels(foo) summary(s5 <- glm(sudi ~ dhb, data = foo, family = binomial(link = "log"))) s5b <- glm(sudi ~ 0 + dhb, data = foo, family = binomial(link = "log")) anova(s5, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 785343 11693 #dhb 3 40.704 785340 11652 0.000000007555 *** dhb.ht <- glht(s5, linfct = mcp(dhb = "Tukey")) summary(dhb.ht) plot(dhb.ht) plot(allEffects(s5)) plot(margins(s5)[[1]]) plot(margins(s5b)[[1]]) m <- s5 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.0009417619 0.0008408861 0.001054739 #dhbMidland 1.3098096206 1.0927743020 1.569950208 #dhbCentral 1.0529486731 0.8658995271 1.280403642 #dhbSouthern 0.6041930829 0.4778598717 0.763925374 ### ## Deprivation # addmargins(table(usudi$dep, usudi$sudi, exclude = NULL)) foo <- usudi[usudi$dep != "Unknown",]; foo <- droplevels(foo) summary(s6 <- glm(sudi ~ dep, data = foo, family = binomial(link = "log"))) s6b <- glm(sudi ~ 0 + dep, data = foo, family = binomial(link = "log")) anova(s6, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 785292 11679 #dep 1 206.68 785291 11472 < 0.00000000000000022 *** dep.ht <- glht(s6, linfct = mcp(dep = "Tukey")) summary(dep.ht) plot(dep.ht) plot(allEffects(s6)) m <- s6 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.0006116709 0.0005509314 0.0006791068 #dep09-10 2.9457046786 2.5482970779 3.4050880992 ### ## Maternal age - categorical addmargins(table(usudi$magecat, usudi$sudi, exclude = NULL)) summary(s7 <- glm(sudi ~ magecat, data = usudi, family = binomial(link = "log"))) s7b <- glm(sudi ~ 0 + magecat, data = usudi, family = binomial(link = "log")) anova(s7, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 786133 9453.2 #magecat 5 241.35 786128 9211.8 < 0.00000000000000022 *** magec.ht <- glht(s7, linfct = mcp(magecat = "Tukey")) summary(magec.ht) plot(magec.ht) plot(allEffects(s7)) plot(margins(s7)[[1]]) plot(margins(s7b)[[1]]) m <- s7 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) # Est LL UL #(Intercept) 0.0002588243 0.0001294476 0.0005175067 #magecatunder20 7.8799783067 3.8448787357 16.1498092353 #magecat20-24 5.3165992300 2.6215406609 10.7822959962 #magecat25-29 2.5353559692 1.2412824514 5.1785392462 #magecat30-34 1.3954781331 0.6755309491 2.8827091087 #magecat35-39 1.5037203419 0.7150663625 3.1621888332 ### ## Birth weight - numeric # summary(s8 <- glm(sudi ~ bwr, data = usudi, family = binomial(link = "log"))) anova(s8, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 785565 11596 #bwr 1 239.59 785564 11356 < 0.00000000000000022 *** plot(allEffects(s8)) m <- s8 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) plot(usudi$bwr,usudi$sudi,xlab="Maternal Age",ylab="Probability of survival") # plot with body size on x-axis and survival (0 or 1) on y-axis g=glm(sudi~bwr,family=binomial,usudi) # run a logistic regression model (in this case, generalized linear model with logit link). see ?glm curve(predict(g,data.frame(bwr=x),type="resp"),add=TRUE) # draws a curve based on prediction from logistic regression model points(usudi$bwr,fitted(g),pch=20) # optional: you could skip this draws an invisible set of points of body size survival based on a 'fit' to glm model. pch= changes type of dots. ### ## Maternal age - numeric # summary(s9 <- glm(sudi ~ mage, data = usudi, family = binomial(link = "log"))) anova(s9, test = "Chisq") #Analysis of Deviance Table #Model: binomial, link: log #Response: sudi #Terms added sequentially (first to last) #Df Deviance Resid. Df Resid. Dev Pr(>Chi) #NULL 786133 9453.2 #mage 1 226.93 786132 9226.2 < 0.00000000000000022 *** plot(allEffects(s9)) m <- s9 se <- sqrt(diag(vcov(m))) # table of estimates with 95% CI (tab <- cbind(Est = coef(m), LL = coef(m) - 1.96 * se, UL = coef(m) + 1.96 *se)) exp(tab) plot(usudi$mage,usudi$sudi,xlab="Maternal Age",ylab="Probability of survival") # plot with body size on x-axis and survival (0 or 1) on y-axis g=glm(sudi~mage,family=binomial,usudi) # run a logistic regression model (in this case, generalized linear model with logit link). see ?glm curve(predict(g,data.frame(mage=x),type="resp"),add=TRUE) # draws a curve based on prediction from logistic regression model points(usudi$mage,fitted(g),pch=20) # optional: you could skip this draws an invisible set of points of body size survival based on a 'fit' to glm model. pch= changes type of dots. library(doBy) orderBy(~ AIC, AIC(s1,s2,s3,s4,s5,s6,s7,s8,s9)) t.test(usudi$mage[usudi$sudi != 1], sudi$mage[usudi$sudi == 1]) #Welch Two Sample t-test # #data: usudi$mage[usudi$sudi != 1] and sudi$mage[usudi$sudi == 1] #t = 15.01, df = 574.84, p-value < 2.2e-16 #alternative hypothesis: true difference in means is not equal to 0 #95 percent confidence interval: # 3.357628 4.368594 #sample estimates: # mean of x mean of y #29.22485 25.36174 plot(predict(s1),residuals(s1),col=c("blue","red")[1+sudi]) lines(lowess(predict(s1)[sudi==0],residuals(s1)[Y==0]),col="blue") lines(lowess(predict(s1)[sudi==1],residuals(s1)[Y==1]),col="red") lines(lowess(predict(s1),residuals(s1)),col="black",lwd=2) abline(h=0,lty=2,col="grey") k <- 20 ##20% of the dataset as testdata N <- 100 ##100 Permutations permu <- paste("Permut_",1:N,sep="") AUC_Results <- matrix(NA, 1, N, dimnames=list("AUC",permu)) n <- ncol(usudi) numrows <- nrow(usudi) learnDataSize <- round(numrows*(1-0.01*k)) testDataSize <- numrows-learnDataSize ##loop for (j in 1:N){ cat("calculating",((j/N)*100),"% \n") learnIndex <-sample(nrow(Dataset))[1:learnDataSize] learnData <-Dataset[learnIndex,] testData <-Dataset[-learnIndex,] mg <-glm(formula =yourFormula, family = binomial(link = "logit"),data=learnData) bestmod_cv <-step(mg,direction="backward",trace=0) predicted_cv <-predict(bestmod_cv, newdata=testData, type="response") observed_cv <-testData[,"Y"] AUC_result <-roc.auc(observed_k, predicted_k) AUC_Results[1,j] <-AUC_result$A } set.seed(0) head(tab <- data.frame(Y=as.numeric(runif(100)>0.5), X=rnorm(100))) subs <- sample(c(1:nrow(tab)), round(nrow(tab)*0.66), replace=F) #the 66% of data you want in one sample tab1 <- tab[subs, ] #the one sample tab2 <- tab[!c(1:nrow(tab)) %in% subs, ] #the other sample, which are the data that do not fall in the first sample rlog1 <- glm(Y~X,family=binomial,data=tab1) summary(rlog1) tab2$pred <-predict(rlog1, newdata=tab2, type="response") hist(tab2$pred) library(ROCR) #allows you to make easily ROC's which allows the assessment of your prediction pred <- prediction(tab2$pred, tab2$Y) perf <- performance(pred,"tpr","fpr") plot(perf); abline(0, 1, col="red") slrm <- lrm(sudi ~ bw, data = usudi)
18d8bb799d504b4d71358a3e5d02a12959eca954
94df6050f2a262da23f62dd678ccc4366b7657fc
/plot-car.R
73a507a2ece28a815707cdd6d402550ac5f96fc6
[]
no_license
EkenePhDAVHV/phd-autonomous-cars-frank
29cc2fc608db53d4d060422022dc5019cf6360f0
1daed3425bfad99dac31543fbeb7950e25aa2878
refs/heads/main
2023-04-29T06:02:59.444072
2021-05-23T11:04:07
2021-05-23T11:04:07
357,157,735
0
0
null
null
null
null
UTF-8
R
false
false
526
r
plot-car.R
library("ggplot2") library("dplyr") d = read.csv("AVHV_Main/output/AVHV_Traffic_Lights/av.csv") d$car_name = as.factor(d$car_name) ggplot(d, aes(x=car_name, y=speed)) + geom_point(aes(col=stopping_time, size=reaction_time)) y = d %>% filter(car_name == "GentleCar29") y$id = 1:nrow(y) ggplot(y, aes(x=id, y=speed)) + geom_point(aes(col=stopping_time, size=reaction_time)) y = d %>% filter(car_name == "GentleCar60") y$id = 1:nrow(y) ggplot(y, aes(x=id, y=speed)) + geom_point(aes(col=stopping_time, size=reaction_time))
3bff5fc50a34205c53b723ef8ccbcf82c81e46f3
8a5830837c7181be89ff732663b5e3e6ec500941
/R/wls.R
d79eb3cadf634b6d0e3f73afcefc2ab8d2a89d7c
[]
no_license
svmiller/stevemisc
282c68d00557537c3be10804e782c64ca757d2e0
5bf1d006908463010866f8fec09300272a3538a0
refs/heads/master
2023-05-11T07:46:56.713314
2023-05-05T13:03:12
2023-05-05T13:03:12
95,028,314
11
4
null
2023-01-12T15:55:12
2017-06-21T17:18:49
R
UTF-8
R
false
false
1,643
r
wls.R
#' Get Weighted Least Squares of Your OLS Model #' #' @description \code{wls()} takes an OLS model and re-estimates it using a weighted least squares #' approach. Weighted least squares is often a "textbook" approach to dealing with the presence of #' heteroskedastic standard errors, for which the weighted least squares estimates are compared #' to the OLS estimates of uncertainty to check for consistency or potential inferential implications. #' #' @details The function *should* be robust to potential model specification oddities (e.g. polynomials #' and fixed effects). It also should perform nicely in the presence of missing data, if and only #' if `na.action = na.exclude` is supplied first to the offending OLS model supplied to the function #' for a weighted least squares re-estimation. #' #' @return \code{wls()} returns a new model object that is a weighted least squares re-estimation #' of the OLS model supplied to it. #' #' @author Steven V. Miller #' #' @param mod a fitted OLS model #' #' @examples #' #' M1 <- lm(mpg ~ ., data=mtcars) #' M2 <- wls(M1) #' #' summary(M2) wls <- function(mod) { if (!identical(class(mod), "lm")) { stop("Weighted Least Squares only makes sense in the context of OLS, and this model/object does not appear to be OLS. As of right now, this function works for just those OLS models generated by the lm() function in base R.") } resid <- resid(mod) fitted <- fitted(mod) WM <- lm(abs(resid) ~ fitted, na.action=na.exclude) wts <- 1/(fitted(WM)^2) A <- eval(mod$call$data) A$wts <- 1/(fitted(WM)^2) WLSM <- update(mod, ~., data=A, weights = wts) return(WLSM) }
15bf498ff0c633a88f463d3bf82990d68020f4e6
9b8c5fba0371b4663f5891bc1ba2225181b59692
/Scripts/0_CV_Cutoff.R
016856d01820429d6c57568c066ac6c65be0ea94
[]
no_license
mpark-bioinfo/HF11
30ac03b32e8329791d13fb8b0ad9ab3177906290
25d1f9b4b10bd2ce91dd385612acbbfbb1b85d37
refs/heads/master
2021-01-21T04:44:21.166060
2016-06-15T21:06:22
2016-06-15T21:06:22
54,639,616
0
0
null
null
null
null
UTF-8
R
false
false
1,765
r
0_CV_Cutoff.R
#!/usr/bin/env Rscript # Meeyoung Park, 04/07/2016 # Command: Rscript <filename> <outdir> # Input: "_imputed.csv" # arguments: cut-off, input_file, output_dir # Process: Extract >= CV cutoff # Output: "_CV.csv" #library(dply) args = commandArgs(TRUE) print(args[2]) # Define CV function cutoff <- args[1] # Read input file lipid_data = read.csv(args[2], header = TRUE, sep = ",") #lipid_data = read.csv("../Normalization/SCNPos/SCNPos_imputed.csv", header = TRUE, sep = ",") # Exclude internal standards is_lipid_idx <- grep("IS", lipid_data$Sample) LipidData <- lipid_data[-is_lipid_idx, ] # Calculate CV TestPool <- LipidData[, 50:length(LipidData)] sd_test <- data.frame(apply(TestPool,1, sd)) colnames(sd_test) <- 'SD' mean_test <- data.frame(rowMeans(TestPool)) colnames(mean_test) <- 'Mean' RSD <- data.frame(as.numeric(sd_test$SD)/as.numeric(mean_test$Mean) * 100) colnames(RSD) <- 'CV'; count.10 <- as.numeric(table(RSD$CV <= 10)["TRUE"]) count.20 <- as.numeric(table(RSD$CV <= 20)["TRUE"]) count.30 <- as.numeric(table(RSD$CV <= 30)["TRUE"]) print(dim(RSD)[1]) print(count.10) print(count.20) print(count.30) percentage10 <- count.10/dim(RSD)[1]*100 percentage20 <- count.20/dim(RSD)[1]*100 percentage30 <- count.30/dim(RSD)[1]*100 print(percentage10) print(percentage20) print(percentage30) CV30_idx <- which((as.numeric(RSD$CV) <= cutoff) == TRUE); FinalLipidData <- LipidData[CV30_idx,]; print(dim(FinalLipidData)); # Output file name name_temp <- strsplit(args[1], "/"); tmp <- strsplit(name_temp[[1]][length(name_temp[[1]])], "\\.") ;#Check the position of file name tissue_name <- strsplit(tmp[[1]][1], "\\_"); new_out <- paste(args[3], tissue_name[[1]][1],"_CV",args[1], ".csv", sep=""); write.csv(FinalLipidData, new_out, row.names=FALSE)
6fadbfe62ab85dd839653647b31882268225f2d3
32c187ff42b70be2dfe109328056d001f5c8b245
/webscraping.R
f8400e59eac608d0a3e1ed062ceb3a195ecb43f9
[]
no_license
npowell759/epl-data-analysis
ea36950d1b4a4d67f1e0b5c0a87d51ac192c0f8e
6e1238a2706086b71001d59960e3b4433f3ad13e
refs/heads/master
2023-03-13T09:02:59.627981
2021-03-05T22:43:55
2021-03-05T22:43:55
344,944,218
0
0
null
null
null
null
UTF-8
R
false
false
1,495
r
webscraping.R
library(rvest) library(dplyr) link <- "https://www.transfermarkt.co.uk/premier-league/transfers/wettbewerb/GB1/plus/?saison_id=2019" page <- read_html(link) team <- page %>% html_nodes("h2 a") %>% html_text() %>% .[21:40] expenditure <- page %>% html_nodes(".transfer-einnahmen-ausgaben.redtext") %>% html_text() %>% gsub("\n\t\tExpenditure: £", "", .) %>% gsub("m\t\t\t\t\t\t", "", .) %>% as.numeric(.) income <- page %>% html_nodes(".transfer-einnahmen-ausgaben.greentext") %>% html_text() %>% gsub("\n\t\t\t\t\t\tIncome: £", "", .) %>% gsub("m\t\t\t\t\t\t\t", "", .) %>% gsub("Th.\t\t\t\t\t\t\t", "", .) %>% as.numeric(.) income[income>250] <- income[income>250] / 1000 income <- round(income, digits = 2) net_spend <- income - expenditure largest_in <- page %>% html_nodes(".responsive-table:nth-child(2) tr:nth-child(1) .rechts a") %>% html_text() %>% gsub("£", "", .) %>% gsub("m", "", .) %>% as.numeric(.) largest_out <- page %>% html_nodes(".responsive-table:nth-child(4) tr:nth-child(1) .rechts a") %>% html_text() %>% gsub("£", "", .) %>% gsub("m", "", .) %>% gsub("Loan fee:", "", .) %>% gsub("Th.", "", .) %>% as.numeric(.) largest_out[largest_out > 140] <- largest_out[largest_out > 140] / 1000 largest_out <- round(largest_out, digits = 2) transferspending <- data.frame(team, expenditure, income, net_spend, largest_in, largest_out, stringsAsFactors = FALSE) write.csv(transferspending, "transferdata.csv", fileEncoding="Windows-1252", row.names=FALSE)
6ca97acc5a6c4b4a11e39877c30207e06723b453
a9392f6fa8733c02e536354c8983c3c437202993
/decisionTreeTraffic.R
4c02fd99728cebb359e22864e822c998723e518e
[]
no_license
abhinav-sunderrajan/Work
363a07da0b17c2288ce996b24ca1414ba1c715d8
cd9b19f6ef3960284127538ce1baa27b3bbba671
refs/heads/master
2021-01-10T14:49:38.617861
2016-01-18T03:39:27
2016-01-18T03:39:27
48,419,991
0
0
null
null
null
null
UTF-8
R
false
false
6,776
r
decisionTreeTraffic.R
library(RPostgreSQL) library(ggplot2) require(gridExtra) library("plot3D") require(tree) library(caret) library(partykit) library("reshape2") library(rpart) library(doParallel) set.seed(2123) drv <- dbDriver("PostgreSQL") con <- dbConnect(drv, host='172.25.187.111', port='5432', dbname='abhinav', user='abhinav', password='qwert$$123') times1=c(4500) times2=times1+120 penetrations=c(0.05,0.1,0.2,0.3,0.4,0.5,1.0) method=c("Uniform","P.I.E","Decision Tree") rSquareMat=matrix(0,length(method),length(penetrations)) rownames(rSquareMat)=method colnames(rSquareMat)=penetrations #PIE pieBreaks=c(0.0,583.9831796057117, 1973.3525390927846, 2367.5948374569775, 2489.872751163762, 3261.271616457807, 4071.903598144841, 4834.836586015808, 5531.1844928841365, 5743.1141418953775, 5965.289275894134, 6207.740716730275, 6670.222224060167, 7025.154292404613, 7658.39571005447, 8040.834612543026, 8554.283981491059, 8807.936731242416, 9591.841494223643, 10148.238120947959, 11286.20075545962, 11637.04476230792, 12438.227750939186, 13328.802132638793) par(mfrow=c(1,1)) #Function for different segmentation methods rSquare<-function(distance_split,time1,time2,penetration,splits){ ls_density=numeric() ls_speed=numeric() i=1 for(name_dist in names(distance_split)){ y<-distance_split[[name_dist]] c2<-cut(y$time_stamp,breaks = seq(time1, time2, by = 1),labels = seq(time1, (time2-1), by = 1)) time_split<-split(y,c2) density_vec=numeric() speed_vec=numeric() index=1 for(name_time in names(time_split)){ num_of_vehicles<-length(unique(time_split[[name_time]]$agent_id))*(1/penetration) density_vec[index]=num_of_vehicles*1000.0/(splits[i+1]-splits[i]) speed_vec[index]=mean(time_split[[name_time]]$speed) index=index+1 } ls_density=c(ls_density,density_vec) ls_speed=c(ls_speed,speed_vec) i=i+1 } good=complete.cases(ls_speed,ls_density) ls_speed=ls_speed[good] ls_density=ls_density[good] good=ls_speed>0 ls_speed=ls_speed[good] ls_density=ls_density[good] fit=lm(ls_density~log(ls_speed)) if((time1==4500 & time2==4620) & penetration==1.0){ plot(ls_speed,ls_density,xlab="",ylab="",cex.axis = 1.50,cex.lab=1.5,xlim=c(0,20),ylim=c(0,250)) points(ls_speed, predict(fit), col="red") mtext("Speed (m/s)", 1, 2.5,cex=1.5) mtext("Density (veh/km)", 2, 2.5,cex=1.5) } summary(fit)$r.squared } sdSplits<-function(distance_split,time1,time2,penetration,splits){ ls_density=numeric() ls_speed=numeric() sd_speed_list=list() i=1 for(name_dist in names(distance_split)){ y<-distance_split[[name_dist]] c2<-cut(y$time_stamp,breaks = seq(time1, time2, by = 60),labels = seq(time1, (time2-60), by = 60)) time_split<-split(y,c2) density_vec=numeric() speed_vec=numeric() sd_speed=numeric() index=1 for(name_time in names(time_split)){ num_of_vehicles<-length(unique(time_split[[name_time]]$agent_id))*(1/penetration) density_vec[index]=num_of_vehicles*1000.0/(splits[i+1]-splits[i]) speed_vec[index]=mean(time_split[[name_time]]$speed) sd_speed[index]=sd(time_split[[name_time]]$speed) index=index+1 } ls_density=c(ls_density,density_vec) ls_speed=c(ls_speed,speed_vec) i=i+1 sd_speed_list[[name_dist]]=c(sd_speed_list[[name_dist]],sd_speed) } sd_speed_list } for(i in 1:1){ query<-paste("select distance_along_road as distance,speed,time_stamp,agent_id from semsim_output WHERE iteration_count=47 AND time_stamp >=",times1[i], " AND time_stamp <=",times2[i]," AND distance_along_road<=13000", sep="") data<-dbGetQuery(con,query) set.seed(i) for(penetration in penetrations){ samples=sample(unique(data$agent_id),length(unique(data$agent_id))*penetration,replace=FALSE) training <- subset(data, agent_id %in% samples) #Decision tree rpart modelFit <- rpart(speed ~ distance,data=training,control=rpart.control(maxdepth=14)) #rpart1a <- as.party(modelFit) #plot(rpart1a) index=1 decisonTreeSplit=numeric() for(split in modelFit$splits[,4]){ decisonTreeSplit[index]=split index=index+1 } decisonTreeSplit<-sort(decisonTreeSplit) decisonTreeSplit<-c(0.0,decisonTreeSplit) #CUTS PIE cutPIE<-cut(training$distance,breaks = pieBreaks,labels = pieBreaks[1:length(pieBreaks)-1]) distance_split_pie<-split(training,cutPIE) statPIE=rSquare(distance_split_pie,times1[i],times2[i],penetration,pieBreaks) rSquareMat["P.I.E",toString(penetration)]=rSquareMat["P.I.E",toString(penetration)]+statPIE #CUTS DECISION TREE cutDT<-cut(training$distance,breaks = decisonTreeSplit,labels = decisonTreeSplit[1:length(decisonTreeSplit)-1]) distance_split_dt<-split(training,cutDT) statDT=rSquare(distance_split_dt,times1[i],times2[i],penetration,decisonTreeSplit) rSquareMat["Decision Tree",toString(penetration)]=rSquareMat["Decision Tree",toString(penetration)]+statDT #Uniform cuts segment_length=1000 segment_breaks=seq(0, 13000, by = segment_length) c1<-cut(training$distance,breaks = segment_breaks,labels = seq(0, (13000-segment_length), by = segment_length)) distance_split<-split(training,c1) statUniform=rSquare(distance_split,times1[i],times2[i],penetration,segment_breaks) rSquareMat["Uniform",toString(penetration)]=rSquareMat["Uniform",toString(penetration)]+statUniform print(paste(times1[i],times2[i],penetration,statPIE,statDT,statUniform,sep=" ")) } } rSquareMat=rSquareMat/5.0 #SD Split query<-paste("select distance_along_road as distance,speed,time_stamp,agent_id from semsim_output WHERE iteration_count=50 AND time_stamp >=",4500, " AND time_stamp <=",4620," AND distance_along_road<=13000", sep="") training<-dbGetQuery(con,query) #Decision tree rpart modelFit <- rpart(speed ~ distance,data=training,control=rpart.control(maxdepth=14)) #rpart1a <- as.party(modelFit) #plot(rpart1a) index=1 decisonTreeSplit=numeric() for(split in modelFit$splits[,4]){ decisonTreeSplit[index]=split index=index+1 } decisonTreeSplit<-sort(decisonTreeSplit) decisonTreeSplit<-c(0.0,decisonTreeSplit) cutDT<-cut(training$distance,breaks = decisonTreeSplit,labels = decisonTreeSplit[1:length(decisonTreeSplit)-1]) distance_split_dt<-split(training,cutDT) sdInSplits=sdSplits(distance_split_dt,4500,4620,1.0,decisonTreeSplit) avgSd=sapply(sdInSplits,mean) #Close Database connection dbDisconnect(con)
3c584db7fc3b9388c04acf39ce5fa6865be1df0f
effd148f74a6e0e7e365953b0ee00918c429f67e
/README.rd
8f1ef61b7c4dc21613c185a7938f9ab122255416
[]
no_license
vijaygithub45/DotNetCoreWebapp
bfa6cc73f213bbe42fa3d48d54b013c511355d55
f885cf63cf373b6a82cf097a11dbe1cff4c543a8
refs/heads/main
2023-04-08T10:02:31.917299
2021-04-19T18:20:49
2021-04-19T18:20:49
356,908,330
0
0
null
2021-04-14T11:12:13
2021-04-11T15:41:32
null
UTF-8
R
false
false
19
rd
README.rd
THIS IS TEST REPO
ae2633d7e0f7703e40f4cc8dd2a5ebf6967c8f3f
f8ce1034cef41685ab2387fa42084ac1ee5c96cf
/chapter13/asggdend.R
35294bfe7e26a204de1a8b4b5767370241dad31e
[]
no_license
AnguillaJaponica/RProgramming
ab7e15d99153ebe585745289490e6e3df118a35c
ae1401920b5a52795cffc95bd83922f15761e531
refs/heads/master
2020-11-28T10:07:02.579925
2019-12-23T15:28:23
2019-12-23T15:28:23
229,777,900
0
0
null
null
null
null
UTF-8
R
false
false
584
r
asggdend.R
library(ggplot2) # デンドログラムの設定 dend2 <- dend %>% set("labels_cex", 0.5) %>% hang.dendrogram(hang_height=0.8) %>% set("branches_k_color", value=c("red", "blue", "green"), k=3) %>% set("branches_lwd", 0.5) %>% set("branches_lty", 1) %>% set("nodes_pch", 19) %>% set("nodes_cex", 1) %>% set("nodes_col", "black") # ggplot2パッケージで描画可能なオブジェクトへの変換 ggd <- as.ggdend(dend2) # ggplot2パッケージでの描画 ggplot(ggd, offset_labels = -0.1)
ec55ebfefeb36b9b130139f6bcf3eaa117c16090
ed633d145dfa8b32511b3cb13ba95b822e2559c8
/doc/gpw.R
36734ae7d5a13a7d8923d631430b78b96eb697fd
[]
no_license
wendellopes/rvswf
51a09f034e330fbb7fd58816c3de2b7f7fdba9dc
ee243c3e57c711c3259a76051a88cc670dfe9c4b
refs/heads/master
2020-05-19T19:38:18.987560
2016-09-11T22:57:37
2016-09-11T22:57:37
19,242,694
1
0
null
null
null
null
UTF-8
R
false
false
2,917
r
gpw.R
lambda=.5e-6 ko=2*pi/lambda lmax<-40 #------------------------------------------------------------------------------- NP=200 z<-seq(-lambda,lambda,lambda/(NP-1)) #------------------------------------------------------------------------------- xo<-sample(z,1) # Entre 1:NP yo<-sample(z,1) # Entre 1:NP zo<-sample(z,1) # Entre 1:NP #------------------------------------------------------------------------------- # Calcula os campos - u e k devem ser ortogonais #------------------------------------------------------------------------------- ux<-sample(-10:10,1) uy<-0 uz<-sample(-10:10,1) #------------------------------------------------------------------------------- kx<-0 ky<-sample(-10:10,1) kz<-0 #------------------------------------------------------------------------------- u<-sqrt(Conj(ux)*ux+Conj(uy)*uy+Conj(uz)*uz) k<-sqrt(Conj(kx)*kx+Conj(ky)*ky+Conj(kz)*kz) #------------------------------------------------------------------------------- # NORMALIZATION #------------------------------------------------------------------------------- ux<-ux/u uy<-uy/u uz<-uz/u um<-(ux-1i*uy)/sqrt(2) up<-(ux+1i*uy)/sqrt(2) #------------------------------------------------------------------------------- # NORMALIZATION #------------------------------------------------------------------------------- kx<-kx/k ky<-ky/k kz<-kz/k km<-(kx-1i*ky)/sqrt(2) kp<-(kx+1i*ky)/sqrt(2) #------------------------------------------------------------------------------- # ELECTRIC FIELD #------------------------------------------------------------------------------- E0<-exp(1i*ko*(kx*xo+ky*yo+kz*zo)) #------------------------------------------------------------------------------- Em<-E0*um Ez<-E0*uz Ep<-E0*up #------------------------------------------------------------------------------- # MAGNETIC FIELD #------------------------------------------------------------------------------- Hm<-1i*E0*(km*uz-kz*um) Hz<-1i*E0*(kp*um-km*up) Hp<-1i*E0*(kz*up-kp*uz) #------------------------------------------------------------------------------- v<-vswf.hmp(ko,xo,yo,zo,lmax) G<-vswf.gpw(ko*kx,ko*ky,ko*kz,ux,uy,uz,lmax) #------------------------------------------------------------------------------- Em.pwe<-sum(G$GTE*v$M.m-G$GTM*v$N.m) Ez.pwe<-sum(G$GTE*v$M.z-G$GTM*v$N.z) Ep.pwe<-sum(G$GTE*v$M.p-G$GTM*v$N.p) Hm.pwe<-sum(G$GTM*v$M.m+G$GTE*v$N.m) Hz.pwe<-sum(G$GTM*v$M.z+G$GTE*v$N.z) Hp.pwe<-sum(G$GTM*v$M.p+G$GTE*v$N.p) #------------------------------------------------------------------------------- H.wfd<-c(Hm,Hz,Hp) E.wfd<-c(Em,Ez,Ep) #------------------------------------------------------------------------------- E.pwe<-c(Em.pwe,Ez.pwe,Ep.pwe) H.pwe<-c(Hm.pwe,Hz.pwe,Hp.pwe) print(cbind(E.pwe,H.pwe)) #------------------------------------------------------------------------------- print(cbind(E.wfd,H.wfd)) print(cbind(E.pwe,H.pwe)) #-------------------------------------------------------------------------------
1279a7c547940f0bccc2a2f45341231f7e963111
95b6831a180d9224004ef7b16f73b77874eb7739
/man/wcs.Rd
018951c28b99fe2a1c35f55d948101d505d1f4c8
[]
no_license
Ketua76/SOWAS
13154b3df7abcfd604d4baafdc894b33b89f0bfa
8a1ec6928f7551f4700c8412014fc4ea3a777566
refs/heads/master
2020-05-22T14:09:36.497283
2012-11-01T15:58:08
2012-11-01T15:58:32
null
0
0
null
null
null
null
UTF-8
R
false
false
3,797
rd
wcs.Rd
\name{wcs} \alias{wcs} \title{Wavelet Sample Cross Spectrum} \usage{ wcs(ts1, ts2, s0 = 1, noctave = 5, nvoice = 10, w0 = 2 * pi, sw = 0, tw = 0, swabs = 0, markt = -999, marks = -999, logscale = FALSE, phase = TRUE, plot = TRUE, units = "", device = "screen", file = "wcsp", split = FALSE, color = TRUE, pwidth = 10, pheight = 7, labsc = 1, labtext = "") } \arguments{ \item{ts1}{first time series object to be transformed} \item{ts2}{second time series object to be transformed} \item{s0}{lowest calculated scale in units of the time series} \item{noctave}{number of octaves} \item{nvoice}{number of voices per octave} \item{w0}{time/frequency resolution omega_0} \item{sw}{length of smoothing window in scale direction is 2*sw*nvoice+1} \item{tw}{length of smoothing window in time direction is 2*s*tw+1} \item{swabs}{length of smoothing window in scale direction at scale s is 2*swabs+1} \item{markt}{vector of times to be marked by vertical dotted lines; when set to -999 (default), no lines are plotted.} \item{marks}{vector of scales to be marked by horizontal dotted lines; when set to -999 (default), no lines are plotted.} \item{logscale}{when TRUE, the contours are plotted in logarithmic scale} \item{phase}{TRUE when phase calculation desired} \item{plot}{TRUE when graphical output desired} \item{units}{character string giving units of the data sets. Default: ""} \item{device}{"screen" or "ps"} \item{file}{character string giving filename of graphical output without extension} \item{split}{when TRUE, modulus and phase are splitted into two files; default: FALSE} \item{color}{TRUE (default): color plot, FALSE: gray scale} \item{pwidth}{width of plot in cm} \item{pheight}{height of plot in cm} \item{labsc}{scale of labels, default: 1, for two-column manuscripts: 1.5, for presentations: >2} \item{labtext}{puts a label in upper left corner of the plot} } \value{ modulus matrix of modulus of wavelet sample cross spectrum of dimension [length(intersection of ts1 and ts2)]x[nvoice*noctave+1] phase matrix of phase of wavelet sample cross spectrum, same dimension as modulus s0 lowest calculated scale in units of the time series noctave number of octaves nvoice number of voices per octave w0 time/frequency resolution omega_0 time vector of times of length(intersection of ts1 and ts2) scales vector of scales of length nvoice*noctave+1 critval not used at not used kernel not used } \description{ This funtion estimates the wavelet cross spectrum of two time series objects with the Morlet wavelet. } \details{ \strong{WARNING!} Better do not use this function because it is in general easily misinterpreted! A peak in the wavelet cross sample spectrum appears in the three cases, that either the first processes exhibits a peak, or the second process or both. But it does not tell, what case is observed. \strong{So in general, a peak in the wavelet cross sample spectrum does not imply that the two underlying processes are related in any way.} The function returns an object of type "wt", that might be directly plotted by the plot function. } \examples{ ## data(nao) data(nino3) # wcs mimics peaks of coherent power, where in reality are non to be # found, as wco shows (see FAQs on my homepage) # Thus, never use wcs! :-) wcsp.nao.nino3 <- wcs(nao,nino3,s0=0.5,noctave=5,nvoice=10) wcoh.nao.nino3 <- wco(nao,nino3,s0=0.5,noctave=5,nvoice=10,sw=0.5,arealsiglevel=0) } \author{ D. Maraun } \references{ D. Maraun and J. Kurths, Nonlin. Proc. Geophys. 11: 505-514, 2004 } \seealso{ \code{\link{cwt.ts}}, \code{\link{wsp}}, \code{\link{wco}} } \keyword{cross} \keyword{spectrum} \keyword{wavelet}
797ddfae0b10dcdc28f87cb11fc1efe4fb10f57f
ae7fed84e88e9626f8387cf5dc92b638be3d90fa
/Postwork1_equipo8.R
4f57e77ab8e220eefb0601ec41cc51053b5e3608
[]
no_license
samanthamartinez-chem/Postworks-equipo-8
4744f4207fcd5392b8b1bed76def686b5f76e7c1
e51f999d5cd57577ca4154e5f732c7114f3c3fff
refs/heads/main
2023-02-28T15:02:23.353289
2021-02-05T03:18:39
2021-02-05T03:18:39
336,146,575
0
0
null
null
null
null
UTF-8
R
false
false
2,824
r
Postwork1_equipo8.R
### ### ## POSTWORK 1 ## ### EQUIPO 8 ### # Se trabajará con datos referentes a equipos de la liga española de fútbol # 1) Importando los datos de soccer de la temporada 2019/2020 de la primera # division de la liga española url <- "https://www.football-data.co.uk/mmz4281/1920/SP1.csv" datos_futbol <- read.csv(url) # 2) Extrayendo las columnas con número de goles anotados por equipos locales # (FTHG), y los goles anotados por los equipos visitantes (FTAG). (tabla_goles_locales <- as.data.frame(table(Goles = datos_futbol$FTHG))) #local (tabla_goles_visitantes <- as.data.frame(table(Goles = datos_futbol$FTAG))) #visita LE1920$FTHG # Goles anotados por los equipos que jugaron en casa LE1920$FTAG # Goles anotados por los equipos que jugaron como visitante # 3) Consultando la función table ?table # Esta función sirve para contruir tablas de frecuencia. #### Parte 2 #### # Elaborando tablas de frecuencia relativas para estimar probabilidades. # Aquí puede visualizar la tabla de frecuencias relativas: # Añadiendo la variable "6 goles" y freucuencia 0 para visitantes, para que los # tamaños de los vectores sean iguales (Tuve que volver a 6 factor) x <- data.frame(Goles = factor(6), Freq = c(0)) # Creando una nuevo df para Visitantes que incluye 6 goles (tabla_goles_visitantes <- rbind.data.frame(tabla_goles_visitantes, x)) # Uniendo frecuencias de Locales y Visitantes en un df merge(tabla_goles_locales, tabla_goles_visitantes, by = "Goles") tabla_frecuencias <- merge(tabla_goles_locales, tabla_goles_visitantes, by = "Goles") colnames(tabla_frecuencias) <- c("Goles", "Locales", "Visitantes") # TABLA DE FRECUENCIAS DE AMBOS EQUIPOS tabla_frecuencias #### AHORA SÍ, LAS PROBABILIDADES #### # 1) La probabilidad (marginal) de que el equipo que juega en casa anote # "x" goles (x = 0, 1, 2, ...) # Obteniendo los goles anotados por equipo local, y dividiendo entre el total # de partidos que se puede obtener con el número de filas. (probmar_locales <- (table(datos_futbol$FTHG)/dim(datos_futbol)[1])*100) # 2) La probabilidad (marginal) de que el equipo que juega como visitante # anote "y" goles (y = 0, 1, 2, ...) # Obteniendo los goles anotados por visitantes y dividiendo entre el total de # partidos. (probmar_visitantes <- (table(datos_futbol$FTAG)/dim(datos_futbol)[1])*100) # 3) La probabilidad (conjunta) de que el equipo que juega en casa anote "x" # goles y el equipo que juega como visitante anote "y" goles # (x = 0, 1, 2, ..., y = 0, 1, 2, ...). (prob_conjunta <- (table(datos_futbol$FTHG, datos_futbol$FTAG)/dim(datos_futbol)[1])*100)
449f6eef6294e157b9a296ee6ecccc24da2f7f34
7c76704087a176e740cc6fbe10682e9c963b129b
/R/dockerfile.R
812aac460e6fc93725272bddf6e3319cc6ffc61f
[ "MIT" ]
permissive
uribo/dockerfiler
d20c0fc352daa39f4ecf7f84c613e170c8683173
2f11b0de7889a2cf60024dc52c90d10859a95075
refs/heads/master
2021-03-30T20:42:24.112672
2018-02-28T17:55:32
2018-02-28T17:55:32
124,981,971
1
0
null
2018-03-13T02:43:09
2018-03-13T02:43:09
null
UTF-8
R
false
false
4,145
r
dockerfile.R
#' A Dockerfile template #' #' @return A dockerfile template #' #' @importFrom R6 R6Class #' @export #' #' @examples #' my_dock <- Dockerfile$new() Dockerfile <- R6::R6Class("Dockerfile", public = list( Dockerfile = character(), ## Either from a file, or from a character vector initialize = function(FROM = "rocker/r-base", AS = NULL){ self$Dockerfile <- create_dockerfile("rocker/r-base", AS) }, RUN = function(cmd){ self$Dockerfile <- c(self$Dockerfile, add_run(cmd)) }, ADD = function(from, to, force = TRUE){ self$Dockerfile <- c(self$Dockerfile,add_add(from, to, force)) }, COPY = function(from, to, force = TRUE){ self$Dockerfile <- c(self$Dockerfile,add_copy(from, to, force)) }, WORKDIR = function(where){ self$Dockerfile <- c(self$Dockerfile, add_workdir(where)) }, EXPOSE = function(port){ self$Dockerfile <- c(self$Dockerfile, add_expose(port)) }, VOLUME = function(volume){ self$Dockerfile <- c(self$Dockerfile, add_volume(volume)) }, CMD = function(cmd){ self$Dockerfile <- c(self$Dockerfile, add_cmd(cmd)) }, LABEL = function(key, value){ self$Dockerfile <- c(self$Dockerfile, add_label(key, value)) }, ENV = function(key, value){ self$Dockerfile <- c(self$Dockerfile, add_env(key, value)) }, ENTRYPOINT = function(cmd){ self$Dockerfile <- c(self$Dockerfile, add_entrypoint(cmd)) }, USER = function(user){ self$Dockerfile <- c(self$Dockerfile, add_user(user)) }, ARG = function(arg, ahead = FALSE){ if (ahead) { self$Dockerfile <- c(add_arg(arg), self$Dockerfile) } else { self$Dockerfile <- c(self$Dockerfile,add_arg(arg)) } }, ONBUILD = function(cmd){ self$Dockerfile <- c(self$Dockerfile,add_onbuild(cmd)) }, STOPSIGNAL = function(signal){ self$Dockerfile <- c(self$Dockerfile,add_stopsignal(signal)) }, HEALTHCHECK = function(check){ self$Dockerfile <- c(self$Dockerfile,add_healthcheck(check)) }, SHELL = function(shell){ self$Dockerfile <- c(self$Dockerfile,add_shell(shell)) }, MAINTAINER = function(name, email){ self$Dockerfile <- c(self$Dockerfile,add_maintainer(name, email)) }, print = function(){ cat(self$Dockerfile, sep = '\n') }, write = function(as = "Dockerfile"){ base::write(self$Dockerfile, file = as) }, switch_cmd = function(a,b){ self$Dockerfile <- switch_them(self$Dockerfile, a, b) }, remove_cmd = function(where){ self$Dockerfile <- remove_from(self$Dockerfile, where) } ))
033076a2d38f46b3d4faee121caafada4578a0d6
9d587824ee032d24c700fe0514c382de04e2bf01
/Pexp.R
2cab6a10cde493d2756f50c0296d42c72c8dc1dc
[]
no_license
Kishor-Kc/ProjectK
518ffad9fe67a7ab4e1f763191ff677bb6dae004
0c80b31bdb7f9ad1948a97bfa9e8c5f5700b3398
refs/heads/master
2021-06-28T03:55:51.119083
2021-03-16T16:09:44
2021-03-16T16:09:44
218,608,169
0
0
null
null
null
null
UTF-8
R
false
false
13,996
r
Pexp.R
library(Rmisc) library(httr) library(RCurl) library(ggplot2) library(dplyr) library(Hmisc) library(data.table) library(tidyverse) library(Rmisc) library(RCurl) library(reshape2) library(ggpubr) library(naniar) #P3 P3_complete <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P3_112019_Eim_Weight%26Oocyst_complete.csv" P3_complete <- read.csv(text = getURL(P3_complete)) P3_complete$X <- NULL P3_a <-select(P3_complete, EH_ID, labels, OPG, wloss, dpi, batch, primary) P3_a <-na.omit(P3_a) P3_b <- select(P3_complete, EH_ID, labels, OPG, wloss, dpi, batch, challenge, infHistory) P3_b <- na.omit(P3_b) # for gene expression P3_design <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experimental_design/P3_112019_Eim_design.csv" P3_design <- read.csv(text = getURL(P3_design)) P3_gene <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P3_112019_Eim_CEWE_RTqPCR.csv" P3_gene <- read.csv(text = getURL(P3_gene)) P3_gene$X <-NULL P3_Delta <-"https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P3_112019_Eim_CEWE_qPCR.csv" P3_Delta <- read.csv(text = getURL(P3_Delta)) P3_Delta$X <- NULL P3_DeltaX <- merge(P3_Delta, P3_design) P3_Comp <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P3_112019_Eim_COMPLETE.csv" P3_Comp <- read.csv(text = getURL(P3_Comp)) P3_Comp$X <- NULL P3_geneX <- select(P3_Comp,"CXCR3", "IL.12", "IRG6", "EH_ID") P3_geneX <- na.omit(P3_geneX) Comp1 <- select(P3_Comp,Eim_MC,EH_ID) Comp1 <- na.omit(Comp1) Comp2 <- merge(P3_DeltaX, Comp1) Comp2 <- distinct(Comp2) compX <- merge(P3_gene, P3_Delta, by = "EH_ID") #boxplot for gene expression complete <- merge(compX, Comp1) Comp.long <- reshape(complete, direction = "long", varying = list(names(complete)[2:4]), v.names = "NE", times = c("CXCR3", "IRG6", "IL.12"), timevar = "Target", idvar = "EH_ID") P3_GE <- select(P3_Comp, EH_ID, challenge) P3_GE <-na.omit(P3_GE) P3_GET <- merge(P3_GE, Comp.long) ggplot(P3_GET, aes(x = challenge, y = NE, color = challenge)) + geom_violin() + geom_jitter() + facet_wrap(~Target) + ggtitle("P3 gene expression") + xlab("Eimeria") + ylab("normalised gene expression") OPGx <- select(P3_Comp,"OPG","dpi","challenge", "EH_ID") OPGx <- na.omit(OPGx) OPG1 <- select(P3_Comp, "OPG", "dpi", "primary", "EH_ID") OPG1 <- na.omit(OPG1) # fix problem with 0340,0335 LM_0340 <- subset(P3_b, P3_b$EH_ID == "LM_0340") LM_O335 <- subset(P3_b, P3_b$EH_ID == "LM_0335") ggplot(P3_b, aes(x = dpi, y = OPG, group = EH_ID, col = challenge)) + geom_smooth()+ geom_line()+ theme_bw()+ facet_grid(~challenge) ggplot(P3_a, aes(x = dpi, y = OPG, group = EH_ID, col = primary)) + geom_smooth()+ geom_line()+ theme_bw()+ facet_grid(~primary) ggplot(P3_Comp, aes(x = dpi, y = OPG, group = EH_ID, col = infHistory)) + geom_point()+ geom_line()+ theme_bw()+ facet_wrap(~infHistory) # for p4 experiment P4_complete <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P4_082020_Eim_COMPLETE.csv" P4_complete <- read.csv(text = getURL(P4_complete)) P4_1 <- select(P4_complete, label, dpi, infHistory, wloss, primary, challenge, OPG) P4_2 <- select(P4_1, label, primary) P4_2 <- na.omit(P4_2) P4_3 <- select(P4_1, label, OPG) P4_3 <- na.omit(P4_3) P4_4 <- select(P4_1,label, dpi, wloss) P4_5 <- merge(P4_2, P4_3) P4_6 <- merge(P4_5, P4_4) #plot OOcyst and weightloss for P4a ggplot(P4_6, aes(x = dpi, y = OPG, color = primary)) + geom_point()+ theme_bw()+ facet_grid(~primary) ggplot(P4_6, aes(x = dpi, y = wloss, col = primary)) + geom_smooth()+ geom_line()+ theme_bw()+ facet_grid(~primary) # plot OOcyst and weightloss for P4b P4_7 <- select(P4_1, label, challenge) P4_7 <- na.omit(P4_7) P4_8 <- select(P4_1, label, OPG) P4_8 <- na.omit(P4_8) P4_9 <- select(P4_1,label, dpi, wloss) P4_10 <- merge(P4_7, P4_8) P4_11 <- merge(P4_10, P4_9) ggplot(P4_11, aes(x = dpi, y = OPG, color = challenge)) + geom_point()+ theme_bw()+ facet_grid(~challenge) ggplot(P4_11, aes(x = dpi, y = wloss, col = challenge)) + geom_smooth()+ geom_line()+ theme_bw()+ facet_grid(~challenge) #add infHis P4_12 <- select(P4_1, label, infHistory, challenge) P4_12 <- na.omit(P4_12) P4_13 <- merge(P4_11, P4_12) ggplot(P4_13, aes(x = dpi, y = OPG, col = infHistory)) + geom_point()+ geom_line()+ theme_bw()+ facet_wrap(~infHistory) # let make P3 + P4 together # for primary wloss P3_w1 <- select(P3_Comp,label, EH_ID, dpi, wloss,primary) P3_w1 <- na.omit(P3_w1) P4_w1 <- select(P4_complete,label, EH_ID, dpi, wloss, primary) P4_w1 <- na.omit(P4_w1) Pr_wloss<- rbind(P3_w1, P4_w1) Pr_wloss <- Pr_wloss[-c(195),] ggplot(Pr_wloss, aes(x = dpi, y = wloss, group = primary, col = primary)) + geom_smooth(se = FALSE)+ geom_jitter()+ theme_bw() + ggtitle("Weightloss during primary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E139", "E64", "E88","Eflab","UNI"), values=c("red", "blue", "darkgreen","orange","violet")) ggplot(Pr_wloss, aes(x = dpi, y = wloss, group = primary, col = primary)) + geom_smooth(se = FALSE)+ geom_jitter()+ theme_bw() + facet_grid(~primary) + ggtitle("Weightloss during primary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)") # sec wloss P3_w2 <- select(P3_Comp,label, EH_ID, dpi, wloss,challenge) P3_w2 <- na.omit(P3_w2) P4_w2 <- select(P4_complete,label, EH_ID, dpi, wloss, challenge) P4_w2 <- na.omit(P4_w2) Ch_wloss <- rbind(P3_w2, P4_w2) Ch_wloss <- distinct(Ch_wloss) ggplot(Ch_wloss, aes(x = dpi, y = wloss, group = challenge, col = challenge )) + geom_smooth(se = FALSE)+ geom_jitter()+ theme_bw() + ggtitle("Weightloss during secondary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) ggplot(Ch_wloss, aes(x = dpi, y = wloss, group = challenge, col = challenge)) + geom_smooth()+ geom_jitter()+ facet_grid(~challenge)+ ggtitle("Weightloss during secondary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) P4_d1 <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experimental_design/P3_112019_Eim_design.csv" P4_d1 <- read.csv(text = getURL(P4_d1)) P4_d2 <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experimental_design/P4a_082020_Eim_design.csv" P4_d2 <- read.csv(text = getURL(P4_d2)) P4_d11 <- select(P4_d1, EH_ID, primary) P4_d21 <- select(P4_d2, EH_ID, primary) P4_D <- rbind(P4_d11, P4_d21) P4_WW <- merge(Ch_wloss, P4_D) ggplot(P4_WW, aes(x = dpi, y = wloss, group = EH_ID, col = primary)) + geom_point()+ geom_line()+ theme_bw()+ facet_grid(~challenge) ggplot(P4_WW, aes(x = dpi, y = wloss, group = challenge, col = challenge)) + geom_smooth(se = FALSE)+ geom_jitter()+ theme_bw()+ facet_grid(~primary) + ggtitle("Weightloss during secondary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)")+ scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) # Ococyst plot #for primary P4_O1 <- select(P4_complete,OPG, dpi, primary, EH_ID) P4_O1 <- na.omit(P4_O1) P4_O1$batch <- c("P4a") P3_O1 <- select(P3_a, OPG, dpi, primary, EH_ID) P3_O1$batch <- c("P3a") OPG_a <- rbind(P4_O1, P3_O1) P3_O2 <- select(P3_b, OPG, dpi, challenge, EH_ID) P3_O2$batch <- c("P3b") P4_O2 <- select(P4_complete,OPG, dpi, challenge, EH_ID) P4_O2 <- na.omit(P4_O2) P4_O2$batch <- c("P4b") Pr_OPG <- rbind(P4_O1, P3_O1) Ch_OPG <- rbind(P4_O2, P3_O2) Inf_His1 <- select(P3_Comp, EH_ID, challenge, infHistory) Inf_His1 <- na.omit(Inf_His1) P3_OPGIH <- merge(Inf_His1,P3_O2) Inf_His2 <- select(P4_complete, EH_ID, challenge, infHistory) Inf_His2 <- na.omit(Inf_His2) P4_OPGIH <- merge(Inf_His2,P4_O2) P_OPGIH <- rbind(P3_OPGIH, P4_OPGIH) P_OPGIH <-na.omit(P_OPGIH) cols <- c("8" = "red", "4" = "blue", "6" = "darkgreen", "10" = "orange","12"="pink") ggplot(Pr_OPG, aes(x = dpi, y = OPG, col = primary)) + geom_smooth()+ geom_point(size=2)+ theme_bw()+ facet_grid(~primary) + ylim(0,5000000) + ggtitle("Oocyst shedding for primary infection ") + ylab("Oocysts per gram (OPG)") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E139", "E64", "E88","Eflab","UNI"), values=c("red", "blue", "darkgreen","orange","violet")) ggplot(P_OPGIH, aes(x = dpi, y = OPG, col = challenge)) + geom_smooth()+ geom_point(size=2)+ theme_bw()+ facet_grid(~challenge) + ylim(0,3000000) + ggtitle("Oocyst shedding for secondary infection ") + ylab("Oocysts per gram (OPG)") + xlab("Day post infection (dpi)")+ scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) ggplot(P_OPGIH, aes(x = dpi, y = OPG, col = challenge)) + geom_smooth()+ geom_point(size=2)+ theme_bw()+ facet_grid(~batch) + ylim(0,3000000) + ggtitle("Oocyst shedding for secondary infection ") + ylab("Oocysts per gram (OPG)") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) ggplot(P_OPGIH, aes(x = dpi, y = OPG, group = challenge, col = batch)) + geom_point()+ geom_line()+ theme_bw()+ facet_wrap(~infHistory) + ggtitle("Oocyst shedding regarding infection history") + ylab("Oocysts per gram (OPG)") + xlab("Day post infection (dpi)") #wlos for sec P_wlossCH <- merge(P_OPGIH,Ch_wloss) ggplot(P_wlossCH, aes(x = dpi, y = wloss, group = challenge, col = challenge)) + geom_smooth()+ geom_jitter()+ facet_grid(~batch)+ ggtitle("Weightloss during secondary infection ") + ylab("Weightloss in percentage") + xlab("Day post infection (dpi)") + scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) # Delta Eim_MC P4_Delta <- select( P4_complete, EH_ID, delta, Eim_MC, infHistory,challenge) P4_Delta <- na.omit(P4_Delta) P4_Delta <- distinct(P4_Delta) P4_Delta$Eim_MC[P4_Delta$Eim_MC == "TRUE"] <- "infected" P4_Delta$Eim_MC[P4_Delta$Eim_MC == "FALSE"] <- "uninfected" P4_Delta$Eim_MC <- as.character(P4_Delta$Eim_MC) P3_Delta <- select(P3_Comp, EH_ID, delta, Eim_MC, infHistory,challenge) P3_Delta <- na.omit(P3_Delta) P3_Delta$Eim_MC <- as.character(P3_Delta$Eim_MC) P3_Delta$Eim_MC[P3_Delta$Eim_MC == "pos"] <- "infected" P3_Delta$Eim_MC[P3_Delta$Eim_MC == "neg"] <- "uninfected" P_Delta <- rbind(P3_Delta, P4_Delta) P_Delta <- distinct(P_Delta) # change LM_0354 uninfected to infected LM_0354 <- subset(P_Delta, P_Delta$EH_ID == "LM_0354") LM_0354$Eim_MC[LM_0354$Eim_MC == "uninfected"] <- "infected" LM337_D <- subset(P_Delta, P_Delta$EH_ID == "LM_0337") LM337_D$Eim_MC[LM337_D$Eim_MC == "infected"] <- "uninfected" LM346_D <- subset(P_Delta, P_Delta$EH_ID == "LM_0346") LM346_D$Eim_MC[LM346_D$Eim_MC == "infected"] <- "uninfected" P_Delta <- P_Delta[-c(9),] P_Delta1 <- P_Delta[-c(13),] P_Delta2 <- P_Delta1[-c(19),] P_Deltax <- rbind(P_Delta2, LM_0354, LM337_D, LM346_D) ggplot(subset(P_Deltax,!is.na(P_Deltax$delta)), aes(x = infHistory, y = delta, color = Eim_MC)) + geom_point()+ theme_bw()+ facet_grid(~challenge) ggplot(P_Deltax, aes(x = Eim_MC, y = delta, color = challenge)) + geom_violin()+ geom_smooth() + geom_jitter()+ theme_bw() ggplot(P_Deltax, aes(x = Eim_MC, y = delta, color = challenge)) + geom_smooth() + geom_jitter()+ ggtitle("Distribution of infection intensity") + xlab("Infection presence") + ylab("Infection intensity") + scale_color_manual(breaks = c("E64", "E88", "UNI"), values=c("blue", "darkgreen","violet")) #try for the average intensity AVG <- subset(P_Deltax, P_Deltax$challenge== "E64") AVG_64 <- mean(AVG[,"delta"], na.rm = TRUE) AVG1 <- subset(P_Deltax, P_Deltax$challenge== "E88") AVG1 <- distinct(AVG1) AVG_88 <- mean(AVG1[,"delta"], na.rm = TRUE) # INFy ELISA result P3_IFN <- "https://raw.githubusercontent.com/derele/Eimeria_Lab/master/data/Experiment_results/P3_112019_Eim_CEWE_ELISA.csv" P3_IFN <- read.csv(text = getURL(P3_IFN)) P3_IFN$X <- NULL P3_E <- merge(P3_IFN,P_Deltax) names(P3_E)[names(P3_E) == "IFNy"] <- "IFNy_CEWE" P3_EE <- select(P3_E, EH_ID, infHistory, delta, challenge, IFNy_CEWE, Eim_MC) P4_E <- select(P4_complete, EH_ID, IFNy_CEWE) P4_E <- na.omit(P4_E) P4_EE <- merge(P_Deltax, P4_E) P_E <- rbind(P3_EE, P4_EE) P_E <- distinct(P_E) ggplot(P_E, aes(y = IFNy_CEWE, x = delta, color= Eim_MC)) + geom_smooth(method="lm")+ geom_jitter()+ theme_bw()+ facet_wrap(~Eim_MC) + stat_cor(aes(label = paste(..rr.label.., ..p.label.., sep = "~`,`~")),label.x = -10, label.y = 750) + ggtitle("Distribution of IFNγ") + ylab("IFNγ from cecum wall") + xlab("Infection intensity") + scale_color_manual(breaks = c("INF", "UNI"), values=c("blue","violet")) P_Inf <- subset(P_E, P_E$Eim_MC == "infected") P_Inf <- distinct(P_Inf) ggplot(P_Inf, aes(x=challenge, y= IFNy_CEWE, color= challenge,main = "Infected" ))+ geom_violin()+ geom_jitter() ggplot(P_Inf, aes(x=delta, y= IFNy_CEWE, color = challenge))+ geom_smooth()+ geom_jitter() + facet_wrap(~challenge) + ggtitle("Presence of IFNγ in infected mice") + ylab("IFNγ from cecum wall") + xlab("Infection intensity")+ scale_color_manual(breaks = c("E64", "E88"), values=c("blue", "darkgreen"))
f49dd5464733c311fc47d3aa57e8bed8fdfe4f02
fafd26bd65bb0afcc6777ce18726dd8b0a5bb1c8
/man/Bus_delays.Rd
34cb02d1b8f45a6383ff06efe2cb44c59d3fc53a
[]
no_license
dtkaplan/SDSdata
50a3814af475c4793e03ec0200abd5d1cec738e9
5f8c9ec18bff5f1a4565cc72f745c4943fd372d9
refs/heads/master
2022-06-27T13:33:10.639497
2022-06-23T20:26:56
2022-06-23T20:26:56
133,568,261
0
1
null
null
null
null
UTF-8
R
false
true
1,804
rd
Bus_delays.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/Bus_delays.R \docType{data} \name{Bus_delays} \alias{Bus_delays} \title{School bus delays in New York City} \format{ A data.frame object with one row for each of 238,266 bus-delay incidents in New York City. \itemize{ \item \code{breakdown_id}: A unique (almost) id number for each incident \item \code{year} the school year of the incident \item \code{route_name} the name of the bus route \item \code{delay_duration} how long (minutes) the delay lasted \item \code{pre_k} whether the bus was for Pre-kindergarden students. If not, it was for elementary school students. (Older students ride the city bus in NYC.) \item \code{reason} an explanation of the cause of the delay \item \code{boro} which of the jurisdictions of the NYC area the delay occurred in. \item \code{n_students} how many student passengers were on the bus when the delay occurred. \item \code{company} the name of the company operating the bus \item \code{date} the date of the incident } } \source{ The data themselves were scraped from the New York City OpenData site, \url{https://data.cityofnewyork.us/Transportation/Bus-Breakdown-and-Delays/ez4e-fazm} on Dec. 13, 2018. The data was brought to the author's attention on a blog written by Nils Ross \url{https://datascienceplus.com/nyc-bus-delays/}. Much of the code used to clean the data was copied from that blog post. } \usage{ data(Bus_delays) } \description{ School bus delays in New York City } \details{ The data have been lightly cleaned from their original form as a CSV file. 305 rows from the original data were removed: those with a number of passengers greater than 48. (These were presumably data-entry errors since a common legal limit for bus passengers is 48.) } \keyword{datasets}
9e0a61fe0a509f8504eeaea54cf075dbe95460c5
c6f9ccacec1e33bd5b132469ea65f7eae8ea1956
/man/subsampleDetections.Rd
85fd51aa55b66de82402ef0e080e29a7ab51ad70
[]
no_license
EricArcher/banter
a2301d738c0ad2572cace2acab16f3058b8f54fb
fa97b2d7ceec5aba51fc9fcce5e9b53a75d352d6
refs/heads/master
2023-03-21T07:39:56.132570
2023-03-14T00:29:58
2023-03-14T00:29:58
127,483,576
7
0
null
null
null
null
UTF-8
R
false
true
1,731
rd
subsampleDetections.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/subsampleDetections.R \name{subsampleDetections} \alias{subsampleDetections} \title{Subsample Detections} \usage{ subsampleDetections(data, n) } \arguments{ \item{data}{a detector data.frame or list of detector data.frames.} \item{n}{a value giving the number (\code{n} >= 1) or fraction (\code{n} between 0 and 1) of detections per event per detector to select. Detections are randomly selected without replacement. If \code{n} is greater than the number of detections in an event, all detections for that event will be retained.} } \value{ a detector data.frame or list of detector data.frames with no more than \code{n} detections per event per detector. } \description{ Extract a random subsample of detections for each event and detector. } \examples{ data(train.data) # initial number of detections per event per detector sapply(train.data$detectors, function(x) table(x$event.id)) # select half of all detectors detect.half <- subsampleDetections(train.data$detectors, 0.5) sapply(detect.half, function(x) table(x$event.id)) # select 20 detections detect.20 <- subsampleDetections(train.data$detectors, 20) sapply(detect.20, function(x) table(x$event.id)) # select 10 detections fro 'ec' detector ec.10 <- subsampleDetections(train.data$detectors$ec, 10) table(ec.10$event.id) } \references{ Rankin, S., Archer, F., Keating, J. L., Oswald, J. N., Oswald, M. , Curtis, A. and Barlow, J. (2017), Acoustic classification of dolphins in the California Current using whistles, echolocation clicks, and burst pulses. Marine Mammal Science 33:520-540. doi:10.1111/mms.12381 } \author{ Eric Archer \email{eric.archer@noaa.gov} }
d0e1a042890b9495844e8b2b5d63415e5dec69ef
265cd044ba3011947844692e0d2d6d071735a64e
/EDMhelper/R/plot_smap.R
4bf1a8dea4f0326b84be784bb51bd3bc41534d0f
[]
no_license
canankarakoc/r_package_EDMhelper
82656008887af66af7dd5ada5d58b96cf6e7aba0
188d2863130a70068b86fcad01cec23792ba2a78
refs/heads/master
2022-04-08T04:42:57.249681
2020-03-02T20:05:46
2020-03-02T20:05:46
183,004,497
0
0
null
null
null
null
UTF-8
R
false
false
1,570
r
plot_smap.R
#' plot_smap #' #' Plot prediction skill vs. theta for the results of an s-map. #' @param smap_out Output from s-map function (or a list of outputs). #' @param predtype Type of prediction metric to use. Can be "rho", "rmse", or "mae". #' @param pname Optional name for the plot. #' @keywords rEDM, s-map #' @return A ggplot object. #' @import ggplot2 #' @export plot_smap<-function(smap_out, predtype="rho", pname=NA) { if(sum(grep("const_p_val", names(smap_out)))==0) { dm<-dim(smap_out[[1]]) lng<-length(smap_out) if(!is.na(pname[1])) { nms<-pname } else { nms<-names(smap_out) } smap_out<-do.call("rbind", smap_out) rownames(smap_out)<-NULL smap_out$variable<-rep(nms, each=dm[1]) } else { smap_out$variable<-pname } variable<-smap_out$variable theta<-smap_out$theta nice_theme <-theme_bw()+ theme(axis.text= element_text(size=12), axis.title = element_text(size=14,face="bold"), legend.text = element_text(size=12), legend.title = element_text(size=12,face="bold"), plot.title = element_text(size=14, face="bold"), strip.text = element_text(size=14, face="bold"), plot.background = element_blank(), panel.grid.major = element_blank(), panel.grid.minor = element_blank()) predtype<-smap_out[,predtype] plot_ccm <- ggplot(smap_out, aes(x=theta, y=predtype))+ geom_line(color="darkred")+ facet_wrap(~variable)+ labs(y="Prediction skill", x="Nonlinearity, Theta")+ nice_theme return(plot_ccm) }
c442df5b6b8ef64b4c7e68a9cddab231ae5a8af3
883a4a0c1eae84485e1d38e1635fcae6ecca1772
/nCompiler/tests/testthat/test-tensorOperations_accessors.R
2ee7e7b8aa0e297ca8459e0bb4895586892662df
[ "BSD-3-Clause" ]
permissive
nimble-dev/nCompiler
6d3a64d55d1ee3df07775e156064bb9b3d2e7df2
392aabaf28806827c7aa7b0b47f535456878bd69
refs/heads/master
2022-10-28T13:58:45.873095
2022-10-05T20:14:58
2022-10-05T20:14:58
174,240,931
56
7
BSD-3-Clause
2022-05-07T00:25:21
2019-03-07T00:15:35
R
UTF-8
R
false
false
11,877
r
test-tensorOperations_accessors.R
context("tensorOperations: special read/write operators") library(nCompiler) library(testthat) library(Matrix) # # test the ability to read/write using nCompiler C++ implementations of diag(), # and related accessor functions # # # generate test data # set.seed(2022) # diagonal matrix creation data nr <- 10 nc <- 3 xv <- runif(n = min(nr, nc)) xv_nr <- runif(n = nr) # random dense matrices X <- matrix(runif(n = 100), nrow = nr) Y <- matrix(runif(n = 100), nrow = nr) # random sparse matrix Xsp <- X Xsp[sample(x = length(Xsp), size = .9 * length(Xsp))] <- 0 Xsp <- Matrix(Xsp, sparse = TRUE) diag(Xsp)[sample(x = nrow(Xsp), size = .1 * nrow(Xsp))] <- 3 # # diag as a creation operator # # Documenting many of R's behaviors for diag(). There are a few other cases # where x may be either a scalar or a vector, expanding upon the ideas # documented here. # # FN. ARGS. NROW NCOL DIAG EIGENIZED # # (x, nrow, ncol) nrow ncol x (x, nrow, ncol) # (nrow, ncol) nrow ncol 1 (1, nrow, ncol) # (x) x x 1 (1, x, x) # (x) length(x) length(x) x (x, length(x), length(x)) # (x, nrow) nrow nrow x (x, nrow, nrow) # (x, ncol) 1 ncol x (x, 1, ncol) # (nrow) nrow nrow 1 (1, nrow, nrow) # (ncol) --- --- - Error, stop processing! diagXRC <- function(x, nrow, ncol) { ans <- diag(x = x, nrow = nrow, ncol = ncol) return(ans) } diagXR <- function(x, nrow) { ans <- diag(x = x, nrow = nrow) return(ans) } diagXC <- function(x, ncol) { ans <- diag(x = x, ncol = ncol) return(ans) } diagRC <- function(nrow, ncol) { ans <- diag(nrow = nrow, ncol = ncol) return(ans) } diagX <- function(x) { ans <- diag(x = x) return(ans) } diagR <- function(nrow) { ans <- diag(nrow = nrow) return(ans) } diagC <- function(ncol) { ans <- diag(ncol = ncol) return(ans) } nDiagXRCv <- nFunction( fun = diagXRC, argTypes = list(x = 'numericVector', nrow = 'integer', ncol = 'integer'), returnType = 'nMatrix' ) nDiagXRC <- nFunction( fun = diagXRC, argTypes = list(x = 'double', nrow = 'integer', ncol = 'integer'), returnType = 'nMatrix' ) nDiagXRv <- nFunction( fun = diagXR, argTypes = list(x = 'numericVector', nrow = 'integer'), returnType = 'nMatrix' ) nDiagXR <- nFunction( fun = diagXR, argTypes = list(x = 'double', nrow = 'integer'), returnType = 'nMatrix' ) nDiagXCv <- nFunction( fun = diagXC, argTypes = list(x = 'numericVector', ncol = 'integer'), returnType = 'nMatrix' ) nDiagXC <- nFunction( fun = diagXC, argTypes = list(x = 'double', ncol = 'integer'), returnType = 'nMatrix' ) nDiagRC <- nFunction( fun = diagRC, argTypes = list(nrow = 'integer', ncol = 'integer'), returnType = 'nMatrix' ) nDiagXv <- nFunction( fun = diagX, argTypes = list(x = 'numericVector'), returnType = 'nMatrix' ) nDiagX <- nFunction( fun = diagX, argTypes = list(x = 'integer'), returnType = 'nMatrix' ) nDiagR <- nFunction( fun = diagR, argTypes = list(nrow = 'integer'), returnType = 'nMatrix' ) nDiagC <- nFunction( fun = diagC, argTypes = list(ncol = 'integer'), returnType = 'nMatrix' ) cDiagXRCv <- nCompile(nDiagXRCv) cDiagXRC <- nCompile(nDiagXRC) cDiagXRv <- nCompile(nDiagXRv) cDiagXR <- nCompile(nDiagXR) cDiagXCv <- nCompile(nDiagXCv) cDiagXC <- nCompile(nDiagXC) cDiagRC <- nCompile(nDiagRC) cDiagXv <- nCompile(nDiagXv) cDiagX <- nCompile(nDiagX) cDiagR <- nCompile(nDiagR) # # creation usage tests # expect_error(nCompile(nDiagC)) # don't support behavior R doesn't support expect_identical(diag(x = xv, nrow = nr, ncol = nc), cDiagXRCv(x = xv, nrow = nr, ncol = nc)) expect_error(cDiagXRCv(x = 4, nrow = nr, ncol = nc)) expect_error(cDiagXRCv(x = runif(nr), nrow = nr, ncol = nc)) expect_identical(cDiagXRC(x = 3, nrow = nr, ncol = nc), diag(x = 3, nrow = nr, ncol = nc)) expect_identical(cDiagXRv(x = xv_nr, nrow = nr), diag(x = xv_nr, nrow = nr)) expect_error(cDiagXRv(x = xv, nrow = nr)) expect_identical(cDiagXR(x = 3, nrow = nr), diag(x = 3, nrow = nr)) expect_identical(cDiagXCv(x = 3, ncol = nc), diag(x = 3, ncol = nc)) expect_identical(cDiagXC(x = 3, ncol = nc), diag(x = 3, ncol = nc)) expect_identical(cDiagRC(nrow = nr, ncol = nc), diag(nrow = nr, ncol = nc)) expect_identical(cDiagXv(x = xv), diag(x = xv)) expect_identical(cDiagX(x = 3), diag(x = 3)) expect_identical(cDiagR(nrow = nr), diag(nrow = nr)) # # sparse creation tests # diagXRC <- function(x, nrow, ncol) { ans <- Diagonal(x = x, nrow = nrow, ncol = ncol) return(ans) } diagXR <- function(x, nrow) { ans <- Diagonal(x = x, nrow = nrow) return(ans) } diagXC <- function(x, ncol) { ans <- Diagonal(x = x, ncol = ncol) return(ans) } diagRC <- function(nrow, ncol) { ans <- Diagonal(nrow = nrow, ncol = ncol) return(ans) } diagX <- function(x) { ans <- Diagonal(x = x) return(ans) } diagR <- function(nrow) { ans <- Diagonal(nrow = nrow) return(ans) } diagC <- function(ncol) { ans <- Diagonal(ncol = ncol) return(ans) } nSpDiagXRCv <- nFunction( fun = diagXRC, argTypes = list(x = 'numericVector', nrow = 'integer', ncol = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXRC <- nFunction( fun = diagXRC, argTypes = list(x = 'double', nrow = 'integer', ncol = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXRv <- nFunction( fun = diagXR, argTypes = list(x = 'numericVector', nrow = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXR <- nFunction( fun = diagXR, argTypes = list(x = 'double', nrow = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXCv <- nFunction( fun = diagXC, argTypes = list(x = 'numericVector', ncol = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXC <- nFunction( fun = diagXC, argTypes = list(x = 'double', ncol = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagRC <- nFunction( fun = diagRC, argTypes = list(nrow = 'integer', ncol = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagXv <- nFunction( fun = diagX, argTypes = list(x = 'numericVector'), returnType = 'nSparseMatrix' ) nSpDiagX <- nFunction( fun = diagX, argTypes = list(x = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagR <- nFunction( fun = diagR, argTypes = list(nrow = 'integer'), returnType = 'nSparseMatrix' ) nSpDiagC <- nFunction( fun = diagC, argTypes = list(ncol = 'integer'), returnType = 'nSparseMatrix' ) cSpDiagXRCv <- nCompile(nSpDiagXRCv) cSpDiagXRC <- nCompile(nSpDiagXRC) cSpDiagXRv <- nCompile(nSpDiagXRv) cSpDiagXR <- nCompile(nSpDiagXR) cSpDiagXCv <- nCompile(nSpDiagXCv) cSpDiagXC <- nCompile(nSpDiagXC) cSpDiagRC <- nCompile(nSpDiagRC) cSpDiagXv <- nCompile(nSpDiagXv) cSpDiagX <- nCompile(nSpDiagX) cSpDiagR <- nCompile(nSpDiagR) expect_error(nCompile(nSpDiagC)) # don't support behavior R doesn't support expect_identical(as(diag(x = xv, nrow = nr, ncol = nc), 'dgCMatrix'), cSpDiagXRCv(x = xv, nrow = nr, ncol = nc)) expect_error(cSpDiagXRCv(x = 4, nrow = nr, ncol = nc)) expect_error(cSpDiagXRCv(x = runif(nr), nrow = nr, ncol = nc)) expect_identical(cSpDiagXRC(x = 3, nrow = nr, ncol = nc), as(diag(x = 3, nrow = nr, ncol = nc), 'dgCMatrix')) expect_identical(cSpDiagXRv(x = xv_nr, nrow = nr), as(diag(x = xv_nr, nrow = nr), 'dgCMatrix')) expect_error(cSpDiagXRv(x = xv, nrow = nr)) expect_identical(cSpDiagXR(x = 3, nrow = nr), as(diag(x = 3, nrow = nr), 'dgCMatrix')) expect_identical(cSpDiagXCv(x = 3, ncol = nc), as(diag(x = 3, ncol = nc), 'dgCMatrix')) expect_identical(cSpDiagXC(x = 3, ncol = nc), as(diag(x = 3, ncol = nc), 'dgCMatrix')) expect_identical(cSpDiagRC(nrow = nr, ncol = nc), as(diag(nrow = nr, ncol = nc), 'dgCMatrix')) expect_identical(cSpDiagXv(x = xv), as(diag(x = xv), 'dgCMatrix')) expect_identical(cSpDiagX(x = 3), as(diag(x = 3), 'dgCMatrix')) expect_identical(cSpDiagR(nrow = nr), as(diag(nrow = nr), 'dgCMatrix')) # # diag as accessor operator # diagExprAccessor <- function(x, y) { ans <- diag(x + y) return(ans) } diagAccessor <- function(x) { ans <- diag(x) return(ans) } nDiagAccessor <- nFunction( fun = diagAccessor, argTypes = list(x = 'nMatrix'), returnType = 'numericVector' ) nDiagExprAccessor <- nFunction( fun = diagExprAccessor, argTypes = list(x = 'nMatrix', y = 'nMatrix'), returnType = 'numericVector' ) nDiagAccessorSp <- nFunction( fun = diagAccessor, argTypes = list(x = 'nSparseMatrix'), returnType = 'numericVector' ) cDiagAccessor <- nCompile(nDiagAccessor) cDiagAccessorSp <- nCompile(nDiagAccessorSp) cDiagExprAccessor <- nCompile(nDiagExprAccessor) # dense accessor expect_identical(as.numeric(cDiagAccessor(X)), diag(X)) # dense accessor of tensor expression objects expect_identical(as.numeric(cDiagExprAccessor(X, Y)), diag(X + Y)) # sparse accessor expect_identical(as.numeric(cDiagAccessorSp(Xsp)), diag(Xsp)) # # diag as assignment operator # diagExprAssignment <- function(x, y, z) { diag(x) <- y + z return(x) } diagAssignment <- function(x, y) { diag(x) <- y return(x) } nDiagExprAssignment <- nFunction( fun = diagExprAssignment, argTypes = list(x = 'nMatrix', y = 'numericVector', z = 'numericVector'), returnType = 'nMatrix' ) nSpDiagExprAssignment <- nFunction( fun = diagExprAssignment, argTypes = list(x = 'nSparseMatrix', y = 'numericVector', z = 'numericVector'), returnType = 'nSparseMatrix' ) nDiagAssignmentv <- nFunction( fun = diagAssignment, argTypes = list(x = 'nMatrix', y = 'numericVector'), returnType = 'nMatrix' ) nDiagAssignment <- nFunction( fun = diagAssignment, argTypes = list(x = 'nMatrix', y = 'double'), returnType = 'nMatrix' ) nSpDiagAssignmentv <- nFunction( fun = diagAssignment, argTypes = list(x = 'nSparseMatrix', y = 'numericVector'), returnType = 'nSparseMatrix' ) nSpDiagAssignment <- nFunction( fun = diagAssignment, argTypes = list(x = 'nSparseMatrix', y = 'double'), returnType = 'nSparseMatrix' ) cDiagExprAssignment <- nCompile(nDiagExprAssignment) cDiagAssignmentv <- nCompile(nDiagAssignmentv) cDiagAssignment <- nCompile(nDiagAssignment) cSpDiagAssignmentv <- nCompile(nSpDiagAssignmentv) cSpDiagAssignment <- nCompile(nSpDiagAssignment) cSpDiagExprAssignment <- nCompile(nSpDiagExprAssignment) # dense assignment via an expression X1 <- X X2 <- X diag(X1) <- diag(Y) + diag(X) expect_identical(X1, cDiagExprAssignment(x = X2, y = diag(Y), z = diag(X))) # dense assignment to vector X1 <- X X2 <- X diag(X1) <- 1:nrow(X) expect_identical(X1, cDiagAssignmentv(x = X2, y = 1:nrow(X))) # dense assignment to constant X1 <- X X2 <- X diag(X1) <- pi expect_identical(X1, cDiagAssignment(x = X2, y = pi)) # sparse assignment to vector X1 <- Xsp X2 <- Xsp diag(X1) = 1:nrow(X1) expect_identical(X1, cSpDiagAssignmentv(x = X2, y = 1:nrow(X1))) # sparse assignment to constant X1 <- Xsp X2 <- Xsp diag(X1) = pi expect_identical(X1, cSpDiagAssignment(x = X2, y = pi)) # sparse assignment via an expression X1 <- Xsp X2 <- Xsp diag(X1) <- diag(Y) + diag(X) expect_identical(X1, cSpDiagExprAssignment(x = X2, y = diag(Y), z = diag(X))) # # usage in composition # d2 <- function(x) { return(diag(diag(x = x))) } d2Sp <- function(x) { return(diag(Diagonal(x = x))) } nD2 <- nFunction( fun = d2, argTypes = list(x = 'integer'), returnType = 'numericVector' ) nD2Sp <- nFunction( fun = d2Sp, argTypes = list(x = 'integer'), returnType = 'numericVector' ) cD2 <- nCompile(nD2) cD2Sp <- nCompile(nD2Sp) expect_equivalent(cD2(x = 5), rep(1,5)) expect_equivalent(cD2Sp(x = 5), rep(1,5))
bae52c7ea499c6f2fff087e052bbb691e306cbb4
bf48bca6d68872cd14b5f26ada56ed23c056beee
/Project2/Problem 4/ActortoID.R
7b78f901abd3733718358ecb5c652ea57cf16647
[]
no_license
coastline1201/ee232e
7186be918c49f0733628e247e848866ae7ceebed
b49cbad655d570dd41e2b62a4b550ca3d0032a70
refs/heads/master
2021-01-10T13:14:52.716399
2016-01-22T05:56:41
2016-01-22T05:56:41
50,161,011
0
3
null
null
null
null
UTF-8
R
false
false
1,266
r
ActortoID.R
# This scripted is used to make a hash table of Actorname -> ActorID # The second part is converting the three new movie’s actor name list to actorID hIDActor = hash() len = length(hActorID) for (i in 1:len) { if(i%%1000==0) { print(i) } hIDActor[hActorID[[paste(i)]]] = i } hIDActress = hash() len = length(hActressID) for (i in 1:len) { if(i%%1000==0) { print(i) } hIDActress[hActressID[[paste(-i)]]] = -i } len1 = length(Node1) Node1Act = vector("numeric",len1) for (i in 1:len1) { if (has.key(Node1[i],hIDActor)) { Node1Act[i] = hIDActor[[Node1[i]]] } if (has.key(Node1[i],hIDActress)) { Node1Act[i] = hIDActress[[Node1[i]]] } } Node1Act = Node1Act[Node1Act!=0] len1 = length(Node2) Node2Act = vector("numeric",len1) for (i in 1:len1) { if (has.key(Node2[i],hIDActor)) { Node2Act[i] = hIDActor[[Node2[i]]] } if (has.key(Node2[i],hIDActress)) { Node2Act[i] = hIDActress[[Node2[i]]] } } Node2Act = Node2Act[Node2Act!=0] len1 = length(Node3) Node3Act = vector("numeric",len1) for (i in 1:len1) { if (has.key(Node3[i],hIDActor)) { Node3Act[i] = hIDActor[[Node3[i]]] } if (has.key(Node3[i],hIDActress)) { Node3Act[i] = hIDActress[[Node3[i]]] } } Node3Act = Node3Act[Node3Act!=0]
0ff5140a92d44c1ce72b2742306fd8edc70b8be6
c983ab9e5564104b93c1301c1d7c153b343a333f
/Chapter 11/Ex11_2_2.R
f082b37e25fb0e45706aec225da7651e262ed658
[]
no_license
SaksheePhade/R-Programming
e638863451f88d7012091e9a34b4fe72f2822919
b5b87a79e3dee2bbb2dd91eacb7abbcdbf446489
refs/heads/master
2021-03-06T10:37:18.710337
2020-11-06T16:25:58
2020-11-06T16:25:58
246,195,251
2
0
null
null
null
null
UTF-8
R
false
false
1,724
r
Ex11_2_2.R
# Page no 549 case_no <- c(1:100) grams <- c(3147 ,2977 ,3119 ,3487 ,4111 ,3572 ,3487 ,3147 , 3345 ,2665 ,1559 ,3799 ,2750 ,3487 ,3317 ,3544 , 3459 ,2807 ,3856 ,3260 ,2183 ,3204 ,3005 ,3090 , 3430 ,3119 ,3912 ,3572 ,3884 ,3090 ,2977 ,3799 , 4054 ,3430 ,3459 ,3827 ,3147 ,3289 ,3629 ,3657 , 3175 ,3232 ,3175 ,3657 ,3600 ,3572 ,709 ,624 , 2778 ,3572 ,3232 ,3317 ,2863 ,3175 ,3317 ,3714 , 2240 ,3345 ,3119 ,2920 ,3430 ,3232 ,3430 ,4139 , 3714 ,1446 ,3147 ,2580 ,3374 ,3941 ,2070 ,3345 , 3600 ,3232 ,3657 ,3487 ,2948 ,2722 ,3771 ,3799 , 1871 ,3260 ,3969 ,3771 ,3600 ,2693 ,3062 ,2693 , 3033 ,3856 ,4111 ,3799 ,3147 ,2920 ,4054 ,2296 , 3402 ,1871 ,4167 ,3402) weeks <- c(40, 41, 38, 38, 39, 41, 40, 41, 38,34, 34, 38, 38, 40, 38, 43, 45, 37,40, 40, 42, 38, 36, 40, 39, 40, 39, 40, 41, 38, 42, 37, 40, 38, 41, 39, 44, 38, 36, 36, 41, 43, 36, 40, 39, 40, 25, 25, 36, 35, 38, 40, 37, 37, 40, 34, 36, 39, 39, 37, 41, 35, 38, 39, 39, 28, 39, 31, 37, 40, 37, 40, 40, 41, 38, 39, 38, 40, 40, 45, 33, 39, 38, 40, 40, 35, 45, 36, 41, 42, 40, 39, 38, 36, 40, 36, 38, 33, 41, 37) smoke <- c(rep(0, 8) ,1 ,rep(0, 6),1 ,rep(0, 4),1 ,0 ,0 , 1 ,rep(0, 9),1 ,0 ,0 ,1 ,0 ,0 ,0 ,1 ,1 ,0 ,1 ,rep(0, 6), rep (0 ,15) ,1 ,1 ,0 ,0 ,0 ,0 ,0 ,0 ,0 ,1 , rep(0 ,10) , 1, rep (0 ,11) ,1 ,0 ,1) data <- data.frame(grams, weeks, smoke) data Regression.eqn <- lm(grams~weeks+smoke) Regression.eqn summary(Regression.eqn) aov <- aov(Regression.eqn) aov summary(aov)
074af1558943a1c7048a3540a1e679de5783207f
c2f9615579153ab57d4e3c80a12fca36988e60fd
/R-script/chap15_(3) 다중공선성(Multicolinearity).R
b246c79eadea4cff1d4d3c70c8a191addbd1d98e
[]
no_license
Elly-bang/R
a0e78a07d01431dd7b774f27a32006035f7646a0
859c87a330b6ec9bba8f881a751318e42029f446
refs/heads/master
2022-11-25T13:35:17.854021
2020-07-13T01:03:37
2020-07-13T01:03:37
277,995,594
0
0
null
null
null
null
UTF-8
R
false
false
2,711
r
chap15_(3) 다중공선성(Multicolinearity).R
################################### # 4. 다중공선성(Multicolinearity) ################################### # - 독립변수 간의 강한 상관관계로 인해서 회귀분석의 결과를 신뢰할 수 없는 현상 # - 생년월일과 나이를 독립변수로 갖는 경우 # - 해결방안 : 강한 상관관계를 갖는 독립변수 제거 # (1) 다중공선성 문제 확인 library(car) fit <- lm(formula=Sepal.Length ~ Sepal.Width+Petal.Length+Petal.Width, data=iris) vif(fit) # Sepal.Width Petal.Length Petal.Width # 1.270815 15.097572 14.234335 sqrt(vif(fit))>2 # root(VIF)가 2 이상인 것은 다중공선성 문제 의심 # Sepal.Width Petal.Length Petal.Width # FALSE TRUE TRUE # (2) iris 변수 간의 상관계수 구하기 cor(iris[,-5]) # 변수간의 상관계수 보기(Species 제외) # # Sepal.Length Sepal.Width Petal.Length Petal.Width # Sepal.Length 1.0000000 -0.1175698 0.8717538 0.8179411 # Sepal.Width -0.1175698 1.0000000 -0.4284401 -0.3661259 # Petal.Length 0.8717538 -0.4284401 1.0000000 0.9628654 (높은 상관관계) # Petal.Width 0.8179411 -0.3661259 0.9628654 1.0000000 #Petal.Length+Petal.Width #x변수 들끼 계수값이 높을 수도 있다. -> 해당 변수 제거(모형 수정) <- Petal.Width # (3) 학습데이터와 검정데이터 분류 : 기계학적 측면 nrow(iris) #150 x <- sample(1:nrow(iris), 0.7*nrow(iris)) # 전체중 70%만 추출 train <- iris[x, ] # 학습데이터 추출 test <- iris[-x, ] # 검정데이터 추출 dim(train) #105 5 dim(test) # 45 5 #모델 검정 # (4) Petal.Width 변수를 제거한 후 회귀분석 result.model<- lm(formula=Sepal.Length ~ Sepal.Width + Petal.Length, data=train) result.model # Call: # lm(formula = Sepal.Length ~ Sepal.Width + Petal.Length, data = train) # # Coefficients: # (Intercept) Sepal.Width Petal.Length # 2.2739 0.5983 0.4622 library(iris) iris_model <- lm(formula = Sepal.Length~Sepal.Width+ Petal.Length, data=train ) summary(iris_model) #(5) model예측치 : test set y_pred <- predict(iris_model,test) y_pred length(y_pred) #45 y_true <- test$Sepal.Length #(6)model평가 :MSE, cor # MSE(표준화o) Error <- y_true - y_pred mse <- mean(Error**2) cat('MSE=',mse) #MSE=0.09154192 #상관계수 r :표준화(X) r <- cor(y_true,y_pred) cat('r=',r) #r=0.9317594 y_pred[1:10] #시각화 평가 plot(y_true, col='blue', type='l', label='y true') points(y_pred, col='red', type='l', label='y pred') #범례 추가 legend("topleft", legend=c,('y true','y pred'), col=c('blue','red'),pch='-')
c6fa38de210a26157a97fc03ee6f0ca346d815ed
2bfd239fd6449e13a5ca753e1c0d0f40407bcfaa
/plot2.R
6c56cd4454c21286a0da371e1697332c089ea8b7
[]
no_license
heber7/Course-Project-2
8d136871ae657da03f9dec33be16863cab1010c7
daf4d381f07e0fc46166314d430c8df1c7899b21
refs/heads/master
2021-01-10T10:39:58.172418
2015-05-24T12:30:58
2015-05-24T12:30:58
36,170,480
0
0
null
null
null
null
UTF-8
R
false
false
503
r
plot2.R
#File Required: NEI <- readRDS("summarySCC_PM25.rds") #-----------------------------------# ## Question 2 - total emmision from PM2.5 Decrrease in Baltimore City from 1999 to 2008 Baltimore <- NEI[NEI$fips == "24510", ] BaltimoreCity <- tapply(Baltimore$Emissions, Baltimore$year, sum) plot(BaltimoreCity, type = "o", pch = 18, cex=2, col = "purple", ylab = "Emissions", xlab = "Year", main = "Emissions: Baltimore City") dev.copy(png,file="plot2.png",width = 480, height = 480) dev.off()
4d4d5e852250bc93b647fe3bed4e178a31c98347
67c56336d7d1236fa57700af759fe75e9454ed44
/util/exechunk.R
c2fff39372bed211d8b3c5332631a15825b4fa25
[]
no_license
actongender/geam-report
b9fbec753a0fca1a4854d60d0abf00a55d3ca65b
163f5b88dacf1ead8cb0397f1ad0e795ffbcb1eb
refs/heads/master
2023-04-28T06:14:36.966608
2021-05-21T06:51:58
2021-05-21T06:51:58
275,156,339
0
0
null
null
null
null
UTF-8
R
false
false
757
r
exechunk.R
#' @title Check several conditions in relation to variable #' #' @description Checks for several conditions related to the execution of a knitr chunk of #' code. Basic check includes if varible exists. #' #' @param variable String. Name of the variable of the corresponding code #' @param data Data frame. Usually df.geam #' #' @return logit #' exechunk <- function(variable, data=NULL){ if (variable %in% skipsubquestion){ return(FALSE) } exec <- TRUE if (is.null(data) & exists("df.geam") ){ cnames <- names(df.geam) } else if (!is.null(data)){ cnames <- names(data) } else { cnames <- "" } exec <- var_exists(variable, cnames, data) return(exec) }
b2bb025ea37d9d6fe4a92174bd1974fb5e8c1575
799475c2ac41f6af3cec0ab5b9ab5136356b884a
/example_LHDdata_multi.R
d84fe8b8d2f95893389d46475ad3ba5775f7de4a
[]
no_license
jtkgithub/trendtests
0f0cf9551140228b88eba7888e1c972aeb6f0a59
b51c1e2692af509212fca9ee70b6baae80fc3d6b
refs/heads/master
2020-04-10T07:10:29.540410
2018-12-12T22:05:59
2018-12-12T22:05:59
160,874,332
0
0
null
null
null
null
ISO-8859-15
R
false
false
3,726
r
example_LHDdata_multi.R
### Examples of how to run the multiple process tests # Clear the workspace rm(list=ls()) # Load the tests source("trendtests_multi.R") ## Read in data # The following data are from: # Kumar and Klefsjø (1992), Reliability analysis of hydraulic systems of LHD machines # using the power law process model, Reliability Engineering and System Safety 35: 217-224. # The data are reported as times between events for machine no 1, 3, 9, 11, 17 and 20. LHD1_xtimes <- c(327, 125, 7, 6, 107, 277, 54, 332, 510, 110, 10, 9, 85, 27, 59, 16, 8, 34, 21, 152, 158, 44, 18) # Times between events LHD1_ftimes <- cumsum(LHD1_xtimes) # Times to event LHD1_tau <- tail(LHD1_ftimes,1) # Use the last event time as censoring time LHD1_ftimes <- LHD1_ftimes[LHD1_ftimes<LHD1_tau] # Remove the censoring time from the event times LHD3_xtimes <- c(637, 40, 397, 36, 54, 53, 97, 63, 216, 118, 125, 25, 4, 101, 184, 167, 81, 46, 18, 32, 219, 405, 20, 248, 140) LHD3_ftimes <- cumsum(LHD3_xtimes) LHD3_tau <- tail(LHD3_ftimes,1) LHD3_ftimes <- LHD3_ftimes[LHD3_ftimes<LHD3_tau] LHD9_xtimes <- c(278, 261, 990, 191, 107, 32, 51, 10, 132, 176, 247, 165, 454, 142, 38, 249, 212, 204, 182, 116, 30, 24, 32, 38, 10, 311, 61) LHD9_ftimes <- cumsum(LHD9_xtimes) LHD9_tau <- tail(LHD9_ftimes,1) LHD9_ftimes <- LHD9_ftimes[LHD9_ftimes<LHD9_tau] LHD11_xtimes <- c(353, 96, 49, 211, 82, 175, 79, 117, 26, 4, 5, 60, 39, 35, 258, 97, 59, 3, 37, 8, 245, 79, 49, 31, 259, 283, 150, 24) LHD11_ftimes <- cumsum(LHD11_xtimes) LHD11_tau <- tail(LHD11_ftimes,1) LHD11_ftimes <- LHD11_ftimes[LHD11_ftimes<LHD11_tau] LHD17_xtimes <- c(401, 36, 18, 159, 341, 171, 24, 350, 72, 303, 34, 45, 324, 2, 70, 57, 103, 11, 5, 3, 144, 80, 53, 84, 218, 122) LHD17_ftimes <- cumsum(LHD17_xtimes) LHD17_tau <- tail(LHD17_ftimes,1) LHD17_ftimes <- LHD17_ftimes[LHD17_ftimes<LHD17_tau] LHD20_xtimes <- c(231, 20, 361, 260, 176, 16, 101, 293, 5, 119, 9, 80, 112, 10, 162, 90, 176, 370, 90, 15, 315, 32, 266) LHD20_ftimes <- cumsum(LHD20_xtimes) LHD20_tau <- tail(LHD20_ftimes,1) LHD20_ftimes <- LHD20_ftimes[LHD20_ftimes<LHD20_tau] # Number of processes considered npros <- 6 # Vector of all censoring times ttauvec <- c(LHD1_tau,LHD3_tau,LHD9_tau,LHD11_tau,LHD17_tau,LHD20_tau) # List of the data set up in the required format SBdatalist <- list(tvec=list(LHD1_ftimes,LHD3_ftimes,LHD9_ftimes,LHD11_ftimes,LHD17_ftimes,LHD20_ftimes), tauvec=ttauvec,m=npros) ## Nelson-Aalen plot of the data nrisk <- npros:1 taus <- sort(ttauvec) etimes <- sort(c(LHD1_ftimes,LHD3_ftimes,LHD9_ftimes,LHD11_ftimes,LHD17_ftimes,LHD20_ftimes)) nrisks <- numeric(length(etimes)) for(i in 1:length(etimes)) nrisks[i] <- head(nrisk[taus>etimes[i]],1) NAest <- cumsum(1/nrisks) plot(etimes,NAest,type="s",xlab="Time (hours)",ylab="Estimated mean cumulative number", lwd=2,col="blue", main="Nelson-Aalen plot") points(etimes,NAest,col="blue") lines(c(head(etimes,1),tail(etimes,1)),c(0,tail(NAest,1)),lty=3,lwd=1,col="red") ## Run the trend tests LRtest_multi(SBdatalist,weights="sqrtNCV",sigma="s") maxtau <- max(ttauvec) # Used to define a in each process in ELR below to correspond to maxtau/2 ELRtest_multi(SBdatalist,weights="sqrtNCV",sigma="s",avec=(1-0.5*ttauvec/maxtau)) CvMtest_multi1(SBdatalist,weights="tau",sigma="s",Npsim=10000) CvMtest_multi2(SBdatalist,weights="tau",sigma="s") ADtest_multi1(SBdatalist,weights="tau",sigma="s",Npsim=10000) ADtest_multi2(SBdatalist,weights="tau",sigma="s") LinRanktest_multi(SBdatalist) GLtest_multi(SBdatalist)
ba487215c62a94cf72b185eca6e80d4a8c98e9eb
d18d09d4184ac140224078017b9cf7ef8f3ae3c1
/R/trump_mein_kampf.R
6f1f0c9794c272f2770f60b18f48737db00484c2
[ "CC0-1.0" ]
permissive
inh2102/TrumpTweetToMeinKampf
3839a4a58215a75d2232550a2604553f4d5c6ea3
a74100d0a4adfd9067cea9bb4b21e332ab4ae376
refs/heads/master
2022-11-27T15:58:51.258186
2020-07-27T23:45:05
2020-07-27T23:45:05
282,809,040
2
0
null
null
null
null
UTF-8
R
false
false
2,701
r
trump_mein_kampf.R
### Just how similar is a given Trump tweet to a line from Hitler's Mein Kampf? Let's find out... ### This script uses four packages: tidyverse (notation), quanteda (text pre-processing and cosine similarity calculation), and rtweet (importing and processing Twitter API data). You'll need to un-comment the install.packages() calls if you haven't installed these packages. #install.packages(tidyverse) #install.packages(quanteda) #install.packages(rtweet) library(tidyverse,quietly=T); library(quanteda,quietly=T); library(rtweet,quietly=T) # As of 7/26/20, Trump has produced 43,137 original tweets (excluding retweets). Let's read those in along with all 10,317 lines of Mein Kampf: trump <- read_csv("all_trump_original_tweets_072620.csv",col_types = cols()) fileName <- 'MeinKampf.txt' MeinKampf <- readChar(fileName, file.info(fileName)$size) %>% strsplit("\\.\\s|\\!\\s|\\?\\s") MeinKampf <- tibble(data.frame(matrix(unlist(MeinKampf), nrow=10317, byrow=T),stringsAsFactors=FALSE)) df <- MeinKampf %>% rename(text = matrix.unlist.MeinKampf...nrow...10317..byrow...T.) # Mein Kampf text pre-processing: hitlercorp <- corpus(df$text) hitlerdfm <- tokens(hitlercorp) %>% tokens_ngrams(n=1:3) %>% dfm(tolower=TRUE,remove_url=TRUE,stem=TRUE,remove_punct=TRUE,remove=c(stopwords("english"))) # Here we define the function: select_trump_tweet <- function() { q1 <- readline(prompt="Which Trump tweet would you like to view? (A number from 1-43137) ") cat(plain_tweets(trump$text)[as.numeric(q1)]) q2 <- readline(prompt="Would you like to use this tweet (Yes or No)? ") if (as.character(tolower(q2))=="yes") { trumpcorp <- corpus(plain_tweets(trump$text)[as.numeric(q1)]) trumpdfm <- tokens(trumpcorp) %>% tokens_ngrams(n=1:10) %>% dfm(tolower=TRUE,remove_url=TRUE,stem=TRUE,remove_punct=TRUE,remove=c(stopwords("english"), "t.co", "https", "rt", "amp", "http", "t.c", "can", "~","RT","realdonaldtrump")) cat("\n\nTrump tweet selected!\n\n") cat("Searching lines from Mein Kampf...\n\n") trump_hitler <- as.data.frame(textstat_simil(hitlerdfm, trumpdfm, margin = "documents",method="cosine")) cat("Done! Top 5 matching results:\n\n") trump_hitler <- as_tibble(trump_hitler[order(-trump_hitler$cosine),]) for (tweet in trump_hitler$document1[1:5]) { cat(paste(which(trump_hitler$document1==tweet),df$text[as.numeric(substr(tweet,5,nchar(tweet)))],"\n\n")) } } else { select_trump_tweet() } } ### Before running the function, if you'd like to view 10 random Trump tweets, you can do so here: sample_n(plain_tweets(trump),10) ## Time to turn a Trump tweet into Mein Kampf. Can you tell the difference? select_trump_tweet()
58e79398f631b5980fdd50efa311ca291b43cd1a
83bfc2ffa4b4e28c1c6ea877c204931980a3e99d
/R_code/simulation_fixed_effect/simulation_tranfromation_fixed_effect_full.R
1c6885f38e859db2d6be86cc237011e47696248c
[]
no_license
wal615/prime_project
0d555626292a713d94700e565363681e2e2e514e
8a85b47ecbcaf4419ca33588fd607019226bf3ca
refs/heads/master
2022-07-04T20:58:33.789355
2020-05-05T20:13:16
2020-05-05T20:13:16
111,431,232
0
0
null
null
null
null
UTF-8
R
false
false
2,541
r
simulation_tranfromation_fixed_effect_full.R
# the following simulation is trying to compare the differnce between the proposed method # and the proposed method on fixed effects. setwd("~/dev/projects/Chen_environmental_study/") source("./R_code/Yang_REML.R") library(sas7bdat) library(MASS) library(tidyverse) library(foreach) library(doRNG) library(doParallel) a=read.sas7bdat("~/dev/projects/Chen_environmental_study/R_code/pcbs1000nomiss.sas7bdat") a=data.matrix(a[,2:35], rownames.force = NA) ## Transromation with full data b <- a b_null_full <- std_fn(b, ncol(b), tran_FUN = null_tran, list(tran = "null", subset = "full")) b_rank_full <- std_fn(b, ncol(b), tran_FUN = rank_tran, list(tran = "rank", subset = "full")) b_quantile_full <- std_fn(b, ncol(b), tran_FUN = norm_quantile_tran, list(tran = "norm_quan", subset = "full")) b_log_full <- std_fn(b, ncol(b), tran_FUN = log_tran, list(tran = "log", subset = "full")) b_sqrt_full <- std_fn(b, ncol(b), tran_FUN = sqrt_tran, list(tran = "sqrt", subset = "full")) b_cat_10_full <- std_fn(b, ncol(b), tran_FUN = categorized_tran, by = 0.1, list(tran = "cate", by = "0.1", subset = "full")) b_cat_5_full <- std_fn(b, ncol(b), tran_FUN = categorized_tran, by = 0.2, list(tran = "cate", by = "0.2", subset = "full")) b_cat_2_full <- std_fn(b, ncol(b), tran_FUN = categorized_tran, by = 0.5, list(tran = "cate", by = "0.5", subset = "full")) data_list_fixed_full <- list(b_null_full = b_null_full, b_rank_full = b_rank_full, b_quantile_full = b_quantile_full, b_log_full = b_log_full, b_sqrt_full = b_sqrt_full, b_cat_10_full = b_cat_10_full, b_cat_5_full = b_cat_5_full, b_cat_2_full = b_cat_2_full) interaction_list <- as.list(rep(1,length(data_list_fixed_full))) interaction_m_list <- as.list(rep(1,length(data_list_fixed_full))) result_list <- mapply(FUN = compare_corr_GCTA, b = data_list_fixed_full, interaction = interaction_list, interaction_m = interaction_m_list, MoreArgs = list(brep = 3, nrep = 100, seed = 123, cores = 3, interm_result = TRUE, interm_result_path = "~/dev/projects/Chen_environmental_study/result/inter_result/"), SIMPLIFY = FALSE) save(result_list, file = "./result/simulation_fixed_full_8tran")
2fc3683167d56ccac9cde141fa85ee197dd38939
db8a43ce4e4d58a57a0a2bb29b63acf6c30b5092
/R/modelmetrics.R
800e2bba79e91cde7e05246c74e7be4167ff6de4
[]
no_license
zhaoxiaohe/MachineShop
ca6fa7d6e7f00ac7d6f8522d50faeec2f4735b2d
85b1ff6a9d7df425d041289856861e75ce596621
refs/heads/master
2020-04-11T06:30:08.059577
2018-12-13T00:45:43
2018-12-13T00:45:43
null
0
0
null
null
null
null
UTF-8
R
false
false
3,977
r
modelmetrics.R
#' Model Performance Metrics #' #' Compute measures of model performance. #' #' @rdname modelmetrics #' #' @param x observed responses or class containing observed and predicted #' responses. #' @param y predicted responses. #' @param metrics function, one or more function names, or list of named #' functions to include in the calculation of performance metrics. #' @param cutoff threshold above which probabilities are classified as success #' for binary responses. #' @param times numeric vector of follow-up times at which survival events #' were predicted. #' @param na.rm logical indicating whether to remove observed or predicted #' responses that are \code{NA} when calculating model metrics. #' @param ... arguments passed from the \code{Resamples} method to the others. #' #' @seealso \code{\link{response}}, \code{\link{predict}}, #' \code{\link{resample}}, \code{\link{metrics}} #' modelmetrics <- function(x, ...) { UseMethod("modelmetrics") } #' @rdname modelmetrics #' #' @examples #' res <- resample(Species ~ ., data = iris, model = GBMModel) #' (metrics <- modelmetrics(res)) #' summary(metrics) #' plot(metrics) #' modelmetrics.Resamples <- function(x, ..., na.rm = TRUE) { control <- x@control if (na.rm) x <- na.omit(x) args <- list(...) args$times <- control@surv_times metrics_by <- by(x, x[c("Model", "Resample")], function(x) { if (nrow(x)) { do.call(modelmetrics, c(list(x$Observed, x$Predicted), args)) } else { NA } }, simplify = FALSE) metrics_list <- tapply(metrics_by, rep(dimnames(metrics_by)$Model, dim(metrics_by)[2]), function(metrics) do.call(rbind, metrics), simplify = FALSE) metrics <- if (length(metrics_list) > 1) { abind(metrics_list, along = 3) } else { metrics_list[[1]] } dimnames(metrics)[[1]] <- dimnames(metrics_by)$Resample ModelMetrics(metrics) } #' @rdname modelmetrics #' modelmetrics.factor <- function(x, y, metrics = c("Accuracy" = MachineShop::accuracy, "Kappa" = MachineShop::kappa2, "ROCAUC" = MachineShop::roc_auc, "Sensitivity" = MachineShop::sensitivity, "Specificity" = MachineShop::specificity, "Brier" = MachineShop::brier), cutoff = 0.5, ...) { metrics <- list2function(metrics) metrics(x, y, cutoff = cutoff) } #' @rdname modelmetrics #' modelmetrics.matrix <- function(x, y, metrics = c("R2" = MachineShop::r2, "RMSE" = MachineShop::rmse, "MAE" = MachineShop::mae), ...) { metrics <- list2function(metrics) metrics(x, y) } #' @rdname modelmetrics #' modelmetrics.numeric <- function(x, y, metrics = c("R2" = MachineShop::r2, "RMSE" = MachineShop::rmse, "MAE" = MachineShop::mae), ...) { metrics <- list2function(metrics) metrics(x, y) } #' @rdname modelmetrics #' #' @examples #' ## Survival response example #' library(survival) #' library(MASS) #' #' fo <- Surv(time, status != 2) ~ sex + age + year + thickness + ulcer #' gbmfit <- fit(fo, data = Melanoma, model = GBMModel) #' #' obs <- response(fo, data = Melanoma) #' pred <- predict(gbmfit, newdata = Melanoma, type = "prob") #' modelmetrics(obs, pred) #' modelmetrics.Surv <- function(x, y, metrics = c("CIndex" = MachineShop::cindex, "ROC" = MachineShop::roc_auc, "Brier" = MachineShop::brier), times = numeric(), ...) { metrics <- list2function(metrics) metrics(x, y, times = times) }
d6751778777e66d757173fbd6037b864c3d79dbc
d0216cc7b6bc2c1694e746a8c176d23405520642
/PGMs/Package_Trial/T3/expMRF/R/SPGM.select.R
8f4a68f7e0d575ea9dab1af8b20b25299175c3bb
[]
no_license
zhandong/XFam
f329947e856afb608eaa3103e02380f982d52bcb
a62925691fa7a5e499d7ad5e76fe903f4cf42914
refs/heads/master
2021-01-19T20:27:37.451797
2014-03-27T14:28:57
2014-03-27T14:28:57
5,893,492
1
2
null
null
null
null
UTF-8
R
false
false
618
r
SPGM.select.R
SPGM.select <- function(X, R, R0=0, N=100,beta=0.05, lmin = 0.01, nlams=20, lambda.path=NULL ,parallel=TRUE,nCpus=4){ require('huge') require('glmnet') if (R < 0){ cat("ERROR: Truncating threshold R should be positive. \n") ghat = NULL return(ghat) } # Generate the matrix with values sublinearly truncated between R (upepr bound) and R0 (lower bound) Xorig <- X X <- round(Bsublin(X, R, R0)) ghat <- LPGM.select(X, method="SPGM", N=N, beta=beta, lmin=lmin, nlams=nlams, lambda.path=lambda.path, parallel=parallel, nCpus=nCpus) if(!is.null(ghat)){ ghat$call = match.call() } return(ghat) }
a3b2704f7213773d676d6e6c4a81ac61b429a0b0
97fbb93b69089b926d299a5d9d44f67888ca4143
/man/dailyprecipNCEP.Rd
45609718b146d3472a03892f7ce6d8a067d2e9b9
[]
no_license
ilyamaclean/microclima
814001c6b36329c477e3497422c21fb2dfd5cf09
8d9900b31c2ff1c58a07e07f71647fde08068e2e
refs/heads/master
2023-08-17T22:27:03.898069
2023-08-04T12:30:03
2023-08-04T12:30:03
139,837,968
30
14
null
2023-04-30T18:44:14
2018-07-05T11:12:32
R
UTF-8
R
false
true
973
rd
dailyprecipNCEP.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/datatools.R \name{dailyprecipNCEP} \alias{dailyprecipNCEP} \title{Obtain daily precipitation from NCEP} \usage{ dailyprecipNCEP(lat, long, tme, reanalysis2 = FALSE) } \arguments{ \item{lat}{the latitude of the location for which data are required} \item{long}{the longitude of the location for which data are required} \item{tme}{a POSIXlt object covering the duration for which data are required. Intervals should be daily. Hourly data are returned irrespective of the time interval of \code{tme}.} \item{reanalysis2}{Logical. Should data be obtained from the Reanalysis II dataset (default) or from Reanalysis I (data prior to 1979).} } \value{ a vector of daily precipitations (mm / day) for the period covered by tme } \description{ Obtain daily precipitation from NCEP } \examples{ tme <- as.POSIXlt(c(1:15) * 24 * 3600, origin = "2015-01-15", tz = 'UTC') dailyprecipNCEP(50, -5, tme) }
c403acac63911df9bf6598fb25dd8c819b2d9270
07c8f492710be978cffd39acec53328bba44e5b8
/auxFunctions/R/adjust_elements_dataframe.R
bf6afde8434f4aad2eb4692898e6ee424c8104b3
[]
no_license
contaratoandressa/pegasus
543957c304c16eaba3a5e3a724a14a16b2455b0c
277c336d8c52c75e902bfdb6f279b070102c5c1c
refs/heads/master
2021-10-02T11:04:20.729333
2021-09-19T00:29:13
2021-09-19T00:29:13
191,631,400
1
0
null
null
null
null
UTF-8
R
false
false
973
r
adjust_elements_dataframe.R
##################################################################### # function that reads a dataframe and adjusts values in factor format ##################################################################### rm(list = ls()) # example dataset <- data.frame(x1 = as.factor(c('teste','Teste0','0TESTE')), x2 = c(1,2,3)) # function #' @param data dataframe #' @param col specific column #' @param name name for change #' @return what the function returns adjust_dataframe <- function(data = dataset, col = 'x1', name = 'teste'){ data[col] <- as.vector(apply(data[col],2,as.character)) data[,col] <- toupper(chartr("ÁÉÍÓÚÀÈÌÒÙÃÕÂÊÔÇáéíóúàèìòùãõâêôç","AEIOUAEIOUAOAEOCaeiouaeiouaoaeoc",data[,col])) data[col][grep("[^:digit:]", data[col])] <- name return(data) } adjust_dataframe() ##################################################################### # end #####################################################################
dcc15de49f6a666827b267a08747970a5a87b599
dc5673ce188e1702e1870bbcdbdec6da34ab55b5
/R_repository/acousaR/R/LF2PGNAPES.R
4b74ce3d28460613c493751176838f85af5bee8e
[]
no_license
saschafassler/acousa
56c04d442e35ae15f19bf0032f89c8ec8e19e5a9
a7d30e66c4aff802643a49d72a41558a4a5efa0c
refs/heads/master
2021-01-10T04:03:18.302152
2015-12-10T14:51:57
2015-12-10T14:51:57
43,646,097
0
0
null
null
null
null
UTF-8
R
false
false
1,953
r
LF2PGNAPES.R
### LF2PGNAPES # load Biology file biology <- read.csv(file="D:/Sascha/Projects/WKEVAL/Dutch HERAS data/PGNAPES format/2014TRI2HERAS_Biology.csv",header=TRUE) # load file with LF data (of fish with just length measurements) lf_data <- read.csv(file="D:/Sascha/Projects/WKEVAL/Dutch HERAS data/acousa format/catchLF_HERAS_2014_HER.csv",header=FALSE) lf_data[is.na(lf_data)] <- 0 # list relevant trawl numbers trawls <- unique(biology$Station) if(unique(biology$Year)==2012) trawls <- trawls[-c(7,8)] # table of aged biological samples: numbers at length lf_biology <- t(table(biology[c("Station","Length")])) # vector of measured lengths lengths <- as.numeric(row.names(lf_biology)) # creating table of additional records (with only length observations) per station for(t in 1: length(trawls)){ station <- trawls[t] Nl_recs <- lf_data[-1,lf_data[1,]==station] - as.numeric(lf_biology[,colnames(lf_biology)==station]) #numbers at length of required records to be added to biological samples reclength <- sum(Nl_recs) #additional records to be added by station station_recs <- data.frame(matrix(ncol = length(colnames(biology)), nrow = reclength)) colnames(station_recs) <- paste(colnames(biology)) # fill additional records table station_recs$Country <- unique(biology$Country) station_recs$Vessel <- unique(biology$Vessel) station_recs$Cruise <- unique(biology$Cruise) station_recs$Station <- station station_recs$StType <- unique(biology$StType) station_recs$Year <- unique(biology$Year) station_recs$Species <- unique(biology$Species) station_recs$Length <- rep(lengths,Nl_recs) station_recs$Recnr <- seq(max(biology$Recnr[which(biology$Station==station)])+1,length=reclength) biology <- rbind(biology,station_recs) } write.csv(biology, file="D:/Sascha/Projects/WKEVAL/Dutch HERAS data/PGNAPES format/2014TRI2HERAS_BiologyALL.csv")
15c0372454555051c12ba3ccde39a3838f2602b7
165ce4507ff54ad5fdd35666af4ebc767faa1231
/man/schema_meta.Rd
0c472f208696a8b06a371c796d53734d9fdbe3f5
[ "MIT" ]
permissive
paulhibbing/AGread
7c99fada4ef1a4c9e17b9a06ae4631e7e82b5f8a
28e9a4946d45cfa0464a10de3ff19f5543a3e6be
refs/heads/master
2022-10-21T12:20:40.233431
2022-09-07T03:04:26
2022-09-07T03:04:26
128,490,460
18
4
NOASSERTION
2021-01-29T21:18:09
2018-04-07T02:47:45
R
UTF-8
R
false
true
364
rd
schema_meta.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/parse_packet_set_SENSOR_SCHEMA.R \name{schema_meta} \alias{schema_meta} \title{Parse metadata from a SENSOR_SCHEMA packet} \usage{ schema_meta(payload) } \arguments{ \item{payload}{raw. the packet payload} } \description{ Parse metadata from a SENSOR_SCHEMA packet } \keyword{internal}
59ee03baab0eb9d7f8e5b42afc8b46fe751ebcb3
f991c879a03a6e5a81141ae59a4fca3a708a55b9
/R/sccosmomcma.R
8d07ad73a1de6d80c2402253ed8a3dc64e50ca12
[ "MIT" ]
permissive
SC-COSMO/sccosmomcma
d1469077b9182d803320ca7ca89b4851a5e81a0f
0e051ae94a2ff0641a9fee09d00097205ea19748
refs/heads/main
2023-04-14T03:12:50.174840
2021-10-11T22:52:25
2021-10-11T22:52:25
371,769,010
3
0
null
null
null
null
UTF-8
R
false
false
388
r
sccosmomcma.R
#' \code{sccosmo} package #' #' The Stanford-CIDE COronavirus Simulation MOdel (SC-COSMO) #' #' See the description of the package on the README on #' \href{https://github.com/feralaes/sccosmo#readme}{GitHub} #' #' @docType package #' @name sccosmo #' @importFrom dplyr %>% filter #' @importFrom rlang .data NULL globalVariables(c(#"df_pop_state_cty_age_us_mex_bra"# ))
28f43672b5774165805635d49034b466b67279a6
b2f61fde194bfcb362b2266da124138efd27d867
/code/dcnf-ankit-optimized/Results/QBFLIB-2018/E1/Database/Amendola-Ricca-Truszczynski/selection-hard/ctrl.e#1.a#3.E#130.A#48.c#.w#7.s#37.asp/ctrl.e#1.a#3.E#130.A#48.c#.w#7.s#37.asp.R
35be74f35e97bedd923bc56cf5b9a4ae0ff654d0
[]
no_license
arey0pushpa/dcnf-autarky
e95fddba85c035e8b229f5fe9ac540b692a4d5c0
a6c9a52236af11d7f7e165a4b25b32c538da1c98
refs/heads/master
2021-06-09T00:56:32.937250
2021-02-19T15:15:23
2021-02-19T15:15:23
136,440,042
0
0
null
null
null
null
UTF-8
R
false
false
91
r
ctrl.e#1.a#3.E#130.A#48.c#.w#7.s#37.asp.R
5ce6f43acd9f5390f7ac98f844906a9a ctrl.e#1.a#3.E#130.A#48.c#.w#7.s#37.asp.qdimacs 7457 21838
008f6963f6778c68b560357d8c252353906ec9d3
0f981f1c09c115bcf6676e561d3d424cc0b2443f
/atacAlign.R
08f740384f686eede2aae148cf704dc54ca8fcd2
[]
no_license
timplab/fragillis
b1b44f84f182adb65132ff7383fa29b166d34308
595c4226541bf8b40c3cb931281a3056414937ea
refs/heads/master
2021-07-17T10:48:01.764770
2018-12-02T20:01:38
2018-12-02T20:01:38
101,221,710
0
0
null
null
null
null
UTF-8
R
false
false
5,901
r
atacAlign.R
library(tidyverse) library(googlesheets) #Code dir is ~/Code/ gs_auth(token = "~/googlesheets_token.rds") fullsheet=gs_title("Fragillis Data") dataloc=gs_read(fullsheet, ws="180608_dat") workdir="/atium/Data/NGS/Aligned/180806_fragillis" outdir="~/Dropbox/timplab_data/fragillis/180806_revision" if (!dir.exists(workdir)) { dir.create(workdir) } if (!dir.exists(outdir)) { dir.create(outdir) } setwd(workdir) if (FALSE) { ##Distribute which set to 3 different computers by changing the starting index for (i in seq(1, dim(dataloc)[1], 3)) { system(paste0("java -jar ~/Code/Trimmomatic-0.38/trimmomatic-0.38.jar PE -phred33 -threads 6 ", dataloc$ill.r1[i], " ", dataloc$ill.r2[i], " /tmp/read1.trim.fq.gz /tmp/read1.unpaired.fq.gz", " /tmp/read2.trim.fq.gz /tmp/read2.unpaired.fq.gz", " ILLUMINACLIP:/home/timp/Code/Trimmomatic-0.38/adapters/NexteraPE-PE.fa:2:30:10", " LEADING:20 TRAILING:20 SLIDINGWINDOW:4:20 MINLEN:40")) system(paste0("bowtie2 -X 2000 -p 10 -t -x /mithril/Data/NGS/Reference/human38/GRCH38 ", "-1 /tmp/read1.trim.fq.gz -2 /tmp/read2.trim.fq.gz ", "2> ", workdir, "/", dataloc$label[i], "_bowtie2.log ", "| samtools view -bS - | samtools sort - -o ", dataloc$label[i], ".bam")) ##started at 8:44AM system(paste0("samtools index ", dataloc$label[i], ".bam")) } } nucchr=paste(c(paste0("chr", 1:22), "chrX", "chrY"), collapse=" ") if (TRUE) { ##remove chrM for (i in 1:dim(dataloc)[1]) { #system(paste0("samtools idxstats ", dataloc$bam[i], " | cut -f 1 | grep -v chrM | xargs samtools view -b ", # dataloc$bam[i], " >", dataloc$label[i], "noM.bam")) system(paste0("samtools view -b ", dataloc$bam[i], " ", nucchr, " >", dataloc$label[i], "noM.bam"), wait=F) } ##remove dups for (i in 1:dim(dataloc)[1]) { system(paste0("java -Xmx8G -jar ~/Code/picard/picard.jar MarkDuplicates ", "REMOVE_DUPLICATES=true INPUT=", dataloc$label[i], "noM.bam", " OUTPUT=", dataloc$label[i], "nodup.bam", " METRICS_FILE=", dataloc$label[i], "_duplicate_metrics.txt")) } for (i in 1:dim(dataloc)[1]) { system(paste0("samtools index ", dataloc$label[i], "nodup.bam"), wait=F) } } ##Generate and share coverage plots if (TRUE) { for (i in 1:dim(dataloc)[1]) { system(paste0("/bin/bash -c ", shQuote(paste0("source activate fragillis; bamCoverage -b ", dataloc$label[i], "nodup.bam -bs 100 -o ", dataloc$label[i], ".bw")))) } for (cond in unique(dataloc$condition)) { just.these=dataloc %>% filter(condition==cond) system(paste0("samtools merge -r ", cond, ".merged.bam ", paste(paste0(just.these$label, "nodup.bam"), collapse=" ")), wait=F) system(paste0("samtools index ", cond, ".merged.bam")) system(paste0("/bin/bash -c ", shQuote(paste0("source activate fragillis; bamCoverage -b ", cond, ".merged.bam -bs 100 -o ", cond, ".bw")))) } } if (TRUE) { ##Merge bam and call peaks on it system(paste0("samtools merge merged.bam ", paste(paste0(dataloc$label, "nodup.bam"), collapse=" "))) ##call peaks system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t merged.bam --broad -n broadmerged --keep-dup all"))), wait=F) system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t merged.bam -n sharpmerged --keep-dup all"))), wait=F) ##per condition for (cond in unique(dataloc$condition)) { just.these=dataloc %>% filter(condition==cond) system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t ", cond, ".merged.bam --broad -n ", cond, ".broadmerged", " --keep-dup all"))), wait=F) system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t ", cond, ".merged.bam -n ", cond, ".sharpmerged", " --keep-dup all"))), wait=F) } ##individ for (i in 1:dim(dataloc)[1]) { system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t ", dataloc$label[i], "nodup.bam --broad -n ", dataloc$label[i], ".broadmerged", " --keep-dup all"))), wait=F) system(paste0("/bin/bash -c ", shQuote(paste0("source activate py27.fragillis; macs2 callpeak ", "--nomodel -g hs -t ", dataloc$label[i], "nodup.bam -n ", dataloc$label[i], ".sharpmerged", " --keep-dup all"))), wait=F) } ##Denny et al command ##macs2 callpeak --nomodle --broad --keep-dup all ##Jawara command ##macs2 callpeak --nomodel -t ${cleaned} -n ${peakout} --nolambda --keep-dup all --shift -75 --extsize 150 -p 0.1 -B -SPMR }
c7c9cb08a0f2aa3c00494e6d03a7060d00a476f2
ffdea92d4315e4363dd4ae673a1a6adf82a761b5
/data/genthat_extracted_code/h2o/examples/h2o.init.Rd.R
ccde17e9c53c2bfe5830fc1376f825a1f1fc0cd3
[]
no_license
surayaaramli/typeRrh
d257ac8905c49123f4ccd4e377ee3dfc84d1636c
66e6996f31961bc8b9aafe1a6a6098327b66bf71
refs/heads/master
2023-05-05T04:05:31.617869
2019-04-25T22:10:06
2019-04-25T22:10:06
null
0
0
null
null
null
null
UTF-8
R
false
false
818
r
h2o.init.Rd.R
library(h2o) ### Name: h2o.init ### Title: Initialize and Connect to H2O ### Aliases: h2o.init ### ** Examples ## Not run: ##D # Try to connect to a local H2O instance that is already running. ##D # If not found, start a local H2O instance from R with the default settings. ##D h2o.init() ##D ##D # Try to connect to a local H2O instance. ##D # If not found, raise an error. ##D h2o.init(startH2O = FALSE) ##D ##D # Try to connect to a local H2O instance that is already running. ##D # If not found, start a local H2O instance from R with 5 gigabytes of memory. ##D h2o.init(max_mem_size = "5g") ##D ##D # Try to connect to a local H2O instance that is already running. ##D # If not found, start a local H2O instance from R that uses 5 gigabytes of memory. ##D h2o.init(max_mem_size = "5g") ## End(Not run)
6f1785f215fd469d2d18b672fb507e6e5b2bb7e3
99e681afa2d7fead8d48034b9536d1360bff04af
/Vaja3/Vaja3.R
b4081e3a47c0d2516f9b842176e26c31864527de
[]
no_license
janpristovnik/Financni_praktikum
9def6419494c9e1fa5dd8b17f13bf415bad537c8
dbed1ea09412b7eb39111ec73d6e779d3fc2a679
refs/heads/master
2021-05-15T09:08:00.548115
2017-12-22T15:43:51
2017-12-22T15:43:51
108,002,774
0
0
null
null
null
null
UTF-8
R
false
false
5,097
r
Vaja3.R
library(combinat) library(Rlab) #naloga 1 #1.a S_0 <- 50 u <- 1.05 d <- 0.95 U <- 5 R <- 0.03 T <- 3 S0 <-c(50.00, 50.00, 50.00, 50.00, 50.00) S1 <-c(52.50, 52.50, 47.50, 47.50, 52.50) S2 <-c(48.88, 55.12, 49.88, 45.12, 49.88) S3 <-c(52.37, 57.88, 47.38, 47.38, 52.37) S4 <-c(49.75, 60.78, 45.01, 49.75, 54.99) S5 <-c(52.24, 63.81, 42.76, 52.24, 57.74) Izplacilo_X <-c(0, 0, 0, 0, 0) Izplacilo_Y <-c(0, 0, 0, 0, 0) tabela1 <- data.frame(S0, S1, S2, S3, S4, S5, Izplacilo_X, Izplacilo_Y) for (i in 1:5){ l = max( max(tabela1[i,(T+1):(U+1)]) - max(tabela1[i, 1:T]), 0) tabela1$Izplacilo_X[i] <- l } for (i in 1:5){ k = max( min(tabela1[i,(T+1):(U+1)]) - min(tabela1[i, 1:T]), 0) tabela1$Izplacilo_Y[i] <- k } #1.b izplacilo <- function(vrsta, T, type = c("call", "put") ) { if (type == "call") { return( max( max(vrsta[(T+1):length(vrsta)]) - max(vrsta[1:T]), 0))} else { return( max( min(vrsta[(T+1):length(vrsta)]) - min(vrsta[1:T]), 0) )} } #2.naloga #a.primer binomski <- function(S0,u,d,U,R,T,type){ q = (1+R-d)/(u-d) razpleti <- hcube(rep(2,U), translation = -1) #drevo stanj 1 pomeni up, 0 down razpleti_1 <- d**(1-razpleti) * u**(razpleti) k <- rowSums(razpleti) #vektor, ki za vsako vrstico pove kolikokrat je up vektor_verjetnosti_koncnih_stanj <- q^k *(1-q)^(U-k) razpleti_1 <- t(apply( razpleti_1, 1, cumprod)) vrednosti <- cbind(S0, S0*razpleti_1) izplacila <- apply(vrednosti, 1, function(x) izplacilo(x,T,type)) E <- sum(izplacila * vektor_verjetnosti_koncnih_stanj) return (E/(1+R)^U) } #2.naloga b primer monte <- function(S0, u, d, U, R, T, type, N){ q = (1+R-d)/(u-d) stanja <- matrix(rbinom(U*N,1,q),N,U) stanja_1 <- d**(1-stanja) * u**(stanja) k <- rowSums(stanja) #vektor, ki za vsako vrstico pove kolikokrat je up vektor_verjetnosti_koncnih_stanj <- q^k *(1-q)^(U-k) stanja_1 <- t(apply(stanja_1, 1, cumprod)) vrednosti <- cbind(S0, S0*stanja_1) izplacila <- apply(vrednosti, 1, function(x) izplacilo(x,T,type)) E= sum(izplacila)/ length(izplacila) return (E/(1+R)^U) } #simuliranje vrednosti monte(60,1.05,0.95,15,0.01,8,"put",10) monte(60,1.05,0.95,15,0.01,8,"put",100) monte(60,1.05,0.95,15,0.01,8,"put",1000) #3.naloga #a.primer N1 <- c() N2 <- c() N3 <- c() M <- 100 for (i in c(1:M)){ N1 <- c(N1,monte(60,1.05,0.95,15,0.01,8,"put",10) ) N2 <- c(N2,monte(60,1.05,0.95,15,0.01,8,"put",100) ) N3 <- c(N3,monte(60,1.05,0.95,15,0.01,8,"put",1000) ) } cena_binomske <- binomski(60,1.05,0.95,15,0.01,8,"put") #cena premije dobljena z binomksim modelom min <- floor(min(c(N1,N2,N3))) max <- ceiling(max(c(N1,N2,N3))) #histogram N1 pov.N1 <- mean(N1) #povprečje vrednosti N1 odklon.N1 <- sqrt(var(N1)) #standardni odklon vrednosti N1 x1_odklon_desno <- cena_binomske + odklon.N1 x1_odklon_levo <- cena_binomske - odklon.N1 histogram1 <-hist(N1,breaks = 20, main = "Monte Carlo: N=10", xlab = "Premija", xlim = c(min, max), col ="yellow") abline(v= pov.N1, col = "green") abline (v = cena_binomske, col = "red", lty = "dashed") arrows(x0 = cena_binomske, y0 = 0, x1= x1_odklon_desno, col= "green", length = 0.1 ) arrows(x0 = cena_binomske, y0 = 0, x1= x1_odklon_levo, col= "green", length = 0.1 ) legend('topright', legend = c('Monte Carlo', 'Analiza modela'), col = c('green', 'red'), cex=0.8, lty=c("solid","dashed")) #histogram N2 pov.N2 <- mean(N2) #povprečje vrednosti N1 odklon.N2 <- sqrt(var(N2)) #standardni odklon vrednosti N1 x2_odklon_desno <- cena_binomske + odklon.N2 x2_odklon_levo <- cena_binomske - odklon.N2 histogram2 <-hist(N2,breaks = 20, main = "Monte Carlo: N=100", xlab = "Premija", xlim = c(min, max), col ="yellow") abline(v= pov.N2, col = "green") abline (v = cena_binomske, col = "red", lty = "dashed") arrows(x0 = cena_binomske, y0 = 0, x1= x2_odklon_desno, col= "green", length = 0.1 ) arrows(x0 = cena_binomske, y0 = 0, x1= x2_odklon_levo, col= "green", length = 0.1 ) legend('topright', legend = c('Monte Carlo', 'Analiza modela'), col = c('green', 'red'), cex=0.8, lty=c("solid","dashed")) #histogram N3 pov.N3 <- mean(N3) #povprečje vrednosti N1 odklon.N3 <- sqrt(var(N3)) #standardni odklon vrednosti N1 x3_odklon_desno <- cena_binomske + odklon.N3 x3_odklon_levo <- cena_binomske - odklon.N3 histogram3 <-hist(N3,breaks = 20, main = "Monte Carlo: N=1000", xlab = "Premija", xlim = c(min, max), col ="yellow") abline(v= pov.N3, col = "green") abline (v = cena_binomske, col = "red", lty = "dashed") arrows(x0 = cena_binomske, y0 = 0, x1= x3_odklon_desno, col= "green", length = 0.1 ) arrows(x0 = cena_binomske, y0 = 0, x1= x3_odklon_levo, col= "green", length = 0.1 ) legend('topright', legend = c('Monte Carlo', 'Analiza modela'), col = c('green', 'red'), cex=0.8, lty=c("solid","dashed"))
22e55a61b99ee150427b4b6a387fbbff4be73ee6
444d2cd04daef3b6ad2d16d288c05ef908481ce7
/Labs5/Labs5-2.R
5f8b2c2be7fa74ed283c545a354aa5062af260d2
[]
no_license
danielreyes9756/ME-LABS
3b25664feaf43d2ceafa6ffc088f47305c822e54
5fc20d1075e928e4ad2102d7d06687d7c720993a
refs/heads/master
2021-01-05T07:34:09.420346
2020-02-16T17:43:53
2020-02-16T17:43:53
240,934,403
0
0
null
null
null
null
UTF-8
R
false
false
568
r
Labs5-2.R
setwd("C:/Users/danie/Desktop") Datos <- read.table("herramienta.txt", dec = ",", header = TRUE) SinHerramientas <- Datos$Sin.herramienta hist(SinHerramientas, labels=TRUE,col="green", density=10, angle = 45, border = "brown") ConHerramientas <- Datos$Con.herramienta hist(ConHerramientas, col="yellow", labels=TRUE, density=10, angle = 45, border = "blue") #Trabajadores distintos t.test(x=SinHerramientas, y=ConHerramientas, alternative="greater",mu=0) #Trabajadores iguales t.test(x=SinHerramientas, y=ConHerramientas, alternative="greater",mu=0, paired = TRUE)
4ff79138ca6e48716b5b2bbcc1244aed5fdc7fbc
d1bf4396480318ae9d7a027bb0001b7d792d70d3
/include/phrank_ml_f.R
47f7bf2b8c11224ee2b314692003c52617760a8f
[]
no_license
mbask/phrank
50ba961710285299323e44cc624318e4809b2c0d
b76c92e3019f661fe92994504411b9996cc1147a
refs/heads/master
2021-01-01T05:16:01.086393
2016-04-20T08:54:29
2016-04-20T08:54:29
56,670,500
0
0
null
null
null
null
UTF-8
R
false
false
4,681
r
phrank_ml_f.R
#' Extract response variable (y) from a formula #' #' @param ml_formula #' #' @return a character vector #' @export #' #' @examples get_response_variable <- function(ml_formula) { formula_terms <- terms(ml_formula) resp_var_name <- attr(formula_terms, "variables")[[attr(formula_terms, "response") + 1]] as.character(resp_var_name) } #' Wrapper around caret::createDataPartition #' #' Pre-defined arguments: #' p = .7, #' list = FALSE, #' times = 1 #' #' @param dataset #' #' @return A list or matrix of row position integers corresponding to the training data #' @export #' @importFrom caret createDataPartition #' #' @examples create_data_partition <- function(dataset) { caret::createDataPartition( dataset$genus_id, p = .7, list = FALSE, times = 1) } #' Subset dataset by extracting train_index rows #' #' @param train_index #' @param dataset #' #' @return filtered dataset #' @export #' #' @examples filter_train_set <- function(train_index, dataset) { dataset[train_index, ] } #' Subset dataset by filtering out train_index rows #' #' @param train_index #' @param dataset #' #' @return filtered dataset #' @export #' #' @examples filter_test_set <- function(train_index, dataset) { dataset[-train_index, ] } #' Wrapper around e1071::tune #' #' Pre-defined arguments: #' method = svm #' ranges = list( #' cost = 2^(0:14), #' gamma = 10^(-8:0)) #' #' @param train_set #' @param svm_formula #' #' @return an object of class "tune" #' @export #' @importFrom e1071 tune #' #' @examples # tune_svm_model <- function(train_set, svm_formula) { # tune( # method = svm, # train.x = svm_formula, # ranges = list( # cost = 2^(0:14), # gamma = 10^(-8:0)), # data = train_set) # } tune_svm_model <- function(train_set, svm_formula) { tune( method = svm, train.x = svm_formula, ranges = list( #cost = 2^(0:14), gamma = seq(0, 1, by = 0.01)), kernel = "radial", data = train_set) } #' Wrapper around e1071::svm #' #' @param train_set #' @param svm_tune_model an object of class "tune" #' @param svm_formula #' #' @return An object of class "svm" #' @export #' @importFrom e1071 svm #' #' @examples # perform_svm_on_train_set <- function(train_set, svm_tune_model, svm_formula) { # svm( # formula = svm_formula, # cost = svm_tune_model$best.parameters$cost, # gamma = svm_tune_model$best.parameters$gamma, # data = train_set) # } #' Wrapper around e1071::predict.svm #' #' @param test_set #' @param svm_model #' @param response_var_name a character vector of the response variable of the svm_model #' #' @return A vector of predicted values #' @export #' @importFrom e1071 predict.svm #' #' @examples predict_svm_on_test_set <- function(test_set, svm_model, response_var_name) { predict( svm_model$best.model, test_set[, colnames(test_set) != response_var_name]) } #' Cross tabulation between the 2 sets of prediction and real factors #' #' It's a wrapper of base::table #' @param prediction_set #' @param real_set #' @param response_var_name #' #' @return #' @export #' #' @examples get_confusion_matrix <- function(prediction_set, real_set, response_var_name) { true_set <- unlist(real_set[, response_var_name]) table( pred = prediction_set, true = true_set) } #' Get accuracy of a classification model from its confusion matrix #' #' Accuracy is defined as the rate of matching row & columns over the total count of elements #' #' @param confusion_matrix #' #' @return #' @export #' #' @examples get_model_accuracy <- function(confusion_matrix) { sum(diag(confusion_matrix) / sum(confusion_matrix)) } get_model_accuracy.randomForest <- function(model) { stopifnot("randomForest" %in% class(model)) get_model_accuracy(model$confusion) } get_model_accuracy.table <- function(model) { stopifnot("table" %in% class(model)) get_model_accuracy(as.matrix(model)) } #' Optimal tree size for a classification random forest #' #' @param m_formula #' @param train_set #' #' @return #' @export #' @importFrom randomForest randomForest #' #' @examples tune_rf_model <- function(train_set, ml_formula) { foo <- randomForest::randomForest( formula = ml_formula, data = train_set) which.min(foo$err.rate[,1]) } #' Title #' #' @param test_set #' @param tree_size #' @param ml_formula #' #' @return #' @export #' @importFrom randomForest randomForest #' #' @examples perform_rf_on_test_set <- function(test_set, tree_size, ml_formula) { randomForest( formula = ml_l$ml_formula, data = ml_l$train_set_l$EVI, ntree = tree_size) }
c82d3ad50c1a6250d0c63c5f53adabddc63b82aa
af5fdc8473585759439e661c55a83628f1a87359
/man/hSearch.Rd
dba132732ab0bee67db693bcc7795c65cb768dca
[]
no_license
ljzhao007/mnspc
f3790be940d0d15c49565eea1eb9c1f62726cc6a
25313d150768afcb692bbf0d0eda20ceef8dce6d
refs/heads/master
2021-03-28T18:27:59.900273
2011-02-21T00:00:00
2011-02-21T00:00:00
null
0
0
null
null
null
null
UTF-8
R
false
false
2,293
rd
hSearch.Rd
\name{hSearch} \alias{hSearch} \title{Searching for \eqn{h} when \eqn{k} is given.} \description{ Given a value of \eqn{k}, this procedure searches for the decision interval \eqn{h} to achieve a pre-specified in-control average run length (ARL). } \usage{ hSearch(f0, arl0, k, Lh, Uh, eh1=0.5, eh2=0.001, istart=0, iter=10000, print.hHistory=TRUE) } \arguments{ \item{f0}{The estimated in-control distribution of the categorized response of the process.} \item{arl0}{The desired in-control ARL.} \item{k}{The allowance constant.} \item{Lh}{Lower bound of the search interval. See details.} \item{Uh}{Upper bound of the search interval. See details.} \item{eh1}{Stopping threshold 1. See details.} \item{eh2}{Stopping threshold 2. See details.} \item{istart}{Steady state start value.} \item{iter}{Number of replications used to compute ARLs throughout the procedure.} \item{print.hHistory}{If \code{TRUE}, the search history will be printed (recommended.)} } \details{ If there is no prior information about the search interval boundaries \code{Lh} and \code{Uh}, they can be omitted. The procedure will first select appropriate decision interval bounds and then search for \eqn{h}. Choosing a very large \code{Uh} may slow down the algorithm substantially and choosing a \code{Uh} that is too small may result in the true \eqn{h} being excluded from the search interval. \code{istart=50} is a good approximation to the steady state start and will not substantially slow down the procedure. Setting \code{istart} to more than 100 is generally not recommended. The stopping rules \code{eh1} and \code{eh2} are explained on pgs. 668-669 of Qiu (2008). \code{iter} should be chosen based on desired precision and speed, although speed is generally not an issue here. This functions calls \code{getARL}. } \value{ A list containing the following: \item{h}{The searched value of \eqn{h}} \item{arl0.actual}{The actual in-control ARL.} } \examples{ ic.dist<-rep(0.125,8) hSearch(f0=ic.dist, arl0=200, k=0.003) hSearch(f0=ic.dist, arl0=200, k=0.967, eh1=0.2, Lh=8, Uh=12, istart=100, iter=5000) } \references{ Qiu, P. (2008) Distribution-free multivariate process control based on log-linear modeling. \emph{IIE Transactions}, 40, 664-677. } \keyword{decision}
6af9eff8d7f752236163ec8b0a0b20a74eb49299
360df3c6d013b7a9423b65d1fac0172bbbcf73ca
/FDA_Pesticide_Glossary/1,2,4-triazole-3-car.R
165ae26840718f95437c711fd6cc85862832201a
[ "MIT" ]
permissive
andrewdefries/andrewdefries.github.io
026aad7bd35d29d60d9746039dd7a516ad6c215f
d84f2c21f06c40b7ec49512a4fb13b4246f92209
refs/heads/master
2016-09-06T01:44:48.290950
2015-05-01T17:19:42
2015-05-01T17:19:42
17,783,203
0
1
null
null
null
null
UTF-8
R
false
false
276
r
1,2,4-triazole-3-car.R
library("knitr") library("rgl") #knit("1,2,4-triazole-3-car.Rmd") #markdownToHTML('1,2,4-triazole-3-car.md', '1,2,4-triazole-3-car.html', options=c("use_xhml")) #system("pandoc -s 1,2,4-triazole-3-car.html -o 1,2,4-triazole-3-car.pdf") knit2html('1,2,4-triazole-3-car.Rmd')
a542b1658526610c92e311ba451d32d3d64b13de
6336fdd99e037619e88b610934dfcca54e063c0f
/man/dmrs.ex.Rd
b2b8955e4e300d5f0dc7102c0e1ca17fbaa94943
[ "MIT" ]
permissive
kdkorthauer/dmrseq
62012f3dfd88e1f3f8acf9b3063198ac0e2531fe
e64d8669f04f0f653ec61d46f8f7071ce9e329ca
refs/heads/master
2023-08-21T16:04:47.550295
2021-11-16T00:56:46
2021-11-16T00:56:46
97,278,089
56
17
MIT
2023-08-04T16:01:36
2017-07-14T22:48:24
R
UTF-8
R
false
true
1,413
rd
dmrs.ex.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/helper_functions.R \docType{data} \name{dmrs.ex} \alias{dmrs.ex} \title{dmrs.ex: Example results of DMRs} \format{a data.frame that contains the results of the inference. The data.frame contains one row for each candidate region, and 10 columns, in the following order: 1. chr = chromosome, 2. start = start basepair position of the region, 3. end = end basepair position of the region, 4. indexStart = the index of the region's first CpG, 5. indexEnd = the index of the region's last CpG, 6. L = the number of CpGs contained in the region, 7. area = the sum of the smoothed beta values 8. beta = the coefficient value for the condition difference, 9. stat = the test statistic for the condition difference, 10. pval = the permutation p-value for the significance of the test statistic, and 11. qval = the q-value for the test statistic (adjustment for multiple comparisons to control false discovery rate).} \source{ Obtained from running the examples in \code{\link{dmrseq}}. A script which executes these steps and constructs the \code{dmrs.ex} object may be found in \file{inst/scripts/get_dmrs.ex.R} } \usage{ data(dmrs.ex) } \description{ Example output from \code{dmrseq} function run on the example dataset \code{BS.chr21}. } \examples{ data(dmrs.ex) } \keyword{datasets}
57bcb5dc9efa382f12642e3f07e659bd22938aec
280578d8d46d88881ad69c8126034a06e08090a8
/data_table_dplyr.R
e7f2cf8e4debe71e5eeec4fa0213af096a2e3188
[ "MIT" ]
permissive
kamilsieklucki/usecases
27aee0ad4e80607800fd4b160d2c1cd1ef9245e8
46b8082c7fc26c52afcc3e2e31bbe3e7dc51d864
refs/heads/master
2021-07-09T08:03:27.550685
2020-07-30T11:34:16
2020-07-30T11:34:16
167,319,679
0
0
null
null
null
null
UTF-8
R
false
false
4,718
r
data_table_dplyr.R
#### https://atrebas.github.io/post/2019-03-03-datatable-dplyr/ #### library(dplyr) library(data.table) df <- mtcars dt <- as.data.table(mtcars) # Slice ---- df %>% slice(5:7) dt[5:7] dt[5:7, ] df %>% slice(-(1:2)) dt[!1:2] dt[-(1:2), ] # Filter ---- df %>% filter(mpg > 20, cyl == 4) dt[mpg > 20 & cyl == 4] # Distinct ---- df %>% distinct() df %>% distinct_at(vars(cyl, gear)) unique(dt) unique(dt, by = c("cyl", "gear")) # zostają wszystkie kolumny więc to nie jest efekt, który był zamierzony unique(dt[, .(cyl, gear)]) # Sort ---- df %>% arrange(desc(mpg), disp) dt[order(-mpg, disp)] # Select ---- df %>% select(mpg, cyl) df %>% select(-(mpg:cyl)) dt[, .(mpg, cyl)] dt[, c("mpg", "cyl")] dt[, !c("mpg", "cyl")] dt[, -c("mpg", "cyl")] # usuwanie zmiennej całkowicie ze zbioru danych dt[, c("mpg", "cyl") := NULL] x <- c("mpg", "cyl") df %>% select(!!x) df %>% select(-!!x) dt[, ..x] dt[, !..x] dt[, -..x] # usuwanie zmiennej całkowicie ze zbioru danych dt[, (x) := NULL] # Summarise ---- df %>% summarise(suma_mpg = sum(mpg), srednia_mpg = mean(mpg)) dt[, .(suma_mpg = sum(mpg), srednia_mpg = mean(mpg))] # Mutate ---- # DT - modyfikuje w miejscu nie trzeba przypisywać do nowej zmiennej, aby otrzymać rezultat df2 <- df %>% mutate(cyl_2 = cyl ^ 2, sqrt_disp = sqrt(disp)) df2 dt[, ':='(cyl_2 = cyl^2, sqrt_disp = sqrt(disp))] dt[, c("cyl_2", "sqrt_disp") := .(cyl^2, sqrt(disp))] dt # Transmute ---- df %>% transmute(cyl_2 = cyl ^ 2) dt[, .(cyl_2 = cyl ^ 2)] # Change values in column ---- df %>% mutate(cyl = replace(cyl, cyl == 4, 0L)) dt[cyl == 4, cyl := 0L] # group by ---- df %>% group_by(cyl, gear) %>% summarise(suma = sum(disp)) dt[, .(suma = sum(disp)), keyby = .(cyl, gear)] # lub by -> The sole difference between by and keyby is that keyby orders the results and creates a key that will allow faster subsetting # count ---- df %>% count(cyl) dt[, .N, keyby = cyl] # add_count ---- df %>% add_count(cyl) dt[, n := .N, by = cyl] # To further manipulate columns, dplyr includes nine functions: the _all, _at, and _if versions of summarise(), mutate(), and transmute(). # With data.table, we use .SD, which is a data.table containing the Subset of Data for each group, excluding the column(s) used in by. So, DT[, .SD] is DT itself and in the expression DT[, .SD, by = V4], .SD contains all the DT columns (except V4) for each values in V4 (see DT[, print(.SD), by = V4]). .SDcols allows to select the columns included in .SD. # summarise_all ---- df %>% summarise_all(max) dt[, lapply(.SD, max)] # summarise_at ---- df %>% group_by(am) %>% summarise_at(vars(cyl, gear), list(min, max)) dt[, c(lapply(.SD, min), lapply(.SD, max)), .SDcols = c("cyl", "gear"), keyby = am] # summarise_if ---- summarise_if(df, nchar(names(df)) == 2, mean) cols <- names(dt)[nchar(names(dt)) == 2] dt[, lapply(.SD, mean), .SDcols = cols] # mutate_all ---- mutate_all(df, as.integer) dt[, lapply(.SD, as.integer)] # analogicznie reszta przypadków # arrange ---- df <- df %>% arrange(cyl) setkey(dt, cyl) setindex(dt, cyl) setorder(dt, mpg) # In data.table, set*() functions modify objects by reference, making these operations fast and memory-efficient. In case this is not a desired behaviour, users can use copy(). The corresponding expressions in dplyr will be less memory-efficient. # set(*) ---- # rename ---- setnames(dt, old = "mpg", new = "test") setnames(dt, old = "test", new = "mpg") # reorder ---- setcolorder(dt, c("carb", "wt", "disp")) # Advanced # Get row number of first (and last) observation by group ---- dt[, .I[c(1, .N)], by = cyl] # suma + poziom wyżej ---- dt[, .(SumV2 = sum(mpg)), keyby = c("cyl", "gear")] rollup(dt, .(SumV2 = sum(mpg)), by = c("cyl", "gear")) # Read and rbind several files ---- rbindlist(lapply(c("DT.csv", "DT.csv"), fread)) # JOINS ---- x <- data.table(Id = c("A", "B", "C", "C"), X1 = c(1L, 3L, 5L, 7L), XY = c("x2", "x4", "x6", "x8"), key = "Id") y <- data.table(Id = c("A", "B", "B", "D"), Y1 = c(1L, 3L, 5L, 7L), XY = c("y1", "y3", "y5", "y7"), key = "Id") # left join x <- y y[x, on = "Id"] left_join(x, y, by = "Id") # right join x -> y x[y, on = "Id"] right_join(x, y, by = "Id") # inner join x[y, on = "Id", nomatch = 0] inner_join(x, y, by = "Id") # full join merge(x, y, all = TRUE, by = "Id") full_join(x, y, by = "Id") # semi_join unique(x[y$Id, on = "Id", nomatch = 0]) semi_join(x, y, by = "Id") # anti_join x[!y, on = "Id"] anti_join(x, y, by = "Id") # Non-equi joins ---- z <- data.table(ID = "C", Z1 = 5:9, Z2 = paste0("z", 5:9)) x[z, on = .(Id == ID, X1 <= Z1)]
a3ec8bb46cbacdb0561ecab6d44ed7523895d57c
ffdea92d4315e4363dd4ae673a1a6adf82a761b5
/data/genthat_extracted_code/zendeskR/examples/getAllOrganizations.Rd.R
b6506acbd5358a219625951c55898e2e40059d72
[]
no_license
surayaaramli/typeRrh
d257ac8905c49123f4ccd4e377ee3dfc84d1636c
66e6996f31961bc8b9aafe1a6a6098327b66bf71
refs/heads/master
2023-05-05T04:05:31.617869
2019-04-25T22:10:06
2019-04-25T22:10:06
null
0
0
null
null
null
null
UTF-8
R
false
false
292
r
getAllOrganizations.Rd.R
library(zendeskR) ### Name: getAllOrganizations ### Title: getAllOrganizations ### Aliases: getAllOrganizations ### Keywords: getAllOrganizations ### ** Examples ## Not run: ##D ## This requires Zendesk authentication ##D organizations <- getAllOrganizations() ## End(Not run)
1a6738f9cce4e9081fe2faac95b7ef3607841824
da41dfabbd285a1a1400d59cbdb7b923bc6d5a35
/algorithm.R
e49809875a09e7041cc1ab43ece6fabecb7d709d
[]
no_license
ferwiner2/FactoryLocation
4e56f2b576b5c22f3341dfcbed2c308d3d4fb5d4
00b9b5cacc193a0b504d72587c3a175c3c0decff
refs/heads/master
2021-01-17T23:45:09.582802
2016-05-21T01:27:10
2016-05-21T01:27:10
null
0
0
null
null
null
null
UTF-8
R
false
false
1,268
r
algorithm.R
install.packages("GA") library("GA") myData <- read.csv(file = "sources.csv") # fitness function euc.dist <- function(x1, x2) sqrt(sum((x1 - x2) ^ 2)) processRow <- function(x, y, row) euc.dist(c(row[1], row[2]), c(x,y)) * row[3] fitnessFunction1 <- function(x) sum( apply(myData, 1, function(row) processRow(x[1], x[2], row))) myClip <- function(x, a, b) { ifelse(x <= a, a, ifelse(x >= b, b, x)) } # this is gareal_raMutation changed to implement # gaussian mutation from https://en.wikipedia.org/wiki/Mutation_(genetic_algorithm) gareal_gaussianMutation <- function(object, parent, ...) { mutate <- parent <- as.vector(object@population[parent,]) n <- length(parent) j <- sample(1:n, size = 1) mutate[j] <- myClip(mutate[j] + rnorm(1), object@min[j], object@max[j]) return(mutate) } solutions <- c() GA <- ga(type="real-valued", fitness = function(x) -fitnessFunction1(x), seed = 0, min = c(0,0), max = c(10, 10), popSize = 20, maxiter = 100, run = 20, selection = function(x) gareal_tourSelection(x, k = 2), pcrossover = 0, mutation = gareal_gaussianMutation, monitor = function(object) solutions <<- append(solutions, object@population))
1285888265953c3069a99dd0f454aaac0ac67937
3b497149a45c9609654d5cbd1da982cbeda530e0
/R/plateaddr.R
bdebbc7ff803e3224d28a8fc5b77683c4ea96c40
[]
no_license
aushev/vautils
f5800a3788bc36b1722884bf7379dcb46327136d
ce770312ecada45305a4e9713a12b8275f1ae12c
refs/heads/master
2023-08-22T10:30:15.931140
2023-08-21T14:36:55
2023-08-21T14:36:55
78,473,057
0
0
null
null
null
null
UTF-8
R
false
false
1,538
r
plateaddr.R
# welltrue('A1a1') = 1 # welltrue('D12h8') = 3072 = 4*12*8*8 welltrue <- function(wellName){ bigRow <- gsub('([a-zA-Z])+.*','\\1',wellName); wellName <- chopLeft(wellName,nchar(bigRow)); bigCol <- gsub('(\\d+).*','\\1',wellName); wellName <- chopLeft(wellName,nchar(bigCol)); smlRow <- gsub('([a-zA-Z]+).*','\\1',wellName); wellName <- chopLeft(wellName,nchar(smlRow)); smlCol <- wellName; bigRowN <- match(tolower(bigRow),letters) - 1L; bigColN <- as.numeric(bigCol) - 1L; smlRowN <- match(tolower(smlRow),letters) - 1L; smlColN <- as.numeric(smlCol); return(bigRowN*12*8*8+bigColN*8*8+smlRowN*8+smlColN); } wellNum2Let <- function(wellNum, base=1L){ # former oaWellNumToAddr() but +1!!! wellNum <- wellNum - base; BigRowN <- wellNum %/% (12*8*8); # 768 BigRow <- LETTERS[BigRowN+1L] remainder <- wellNum %% 768; # wellNum - BigRowN*768; # BigColN <- remainder %/% (8*8); BigCol <- BigColN + 1L; remainder <- remainder %% 64 SmlRowN <- remainder %/% 8; SmlRow <- letters[SmlRowN+1L] SmlCol <- remainder - SmlRowN*8 + 1L; return(paste0(BigRow,BigCol,SmlRow,SmlCol)); } oaWellNumToAddr <- function(wellNum){ # BigRowN <- wellNum %/% (12*8*8); # 768 BigRow <- LETTERS[BigRowN+1L] remainder <- wellNum %% 768; # wellNum - BigRowN*768; # BigColN <- remainder %/% (8*8); BigCol <- BigColN + 1L; remainder <- remainder %% 64 SmlRowN <- remainder %/% 8; SmlRow <- letters[SmlRowN+1L] SmlCol <- remainder - SmlRowN*8 + 1L; return(paste0(BigRow,BigCol,SmlRow,SmlCol)); }
b1ab0327bb5029af41193f857315b58c5de21a2c
59fc6c331e0a4aa604860b899cba2a9331311439
/man/dsumlogchisq.Rd
64591874849579ea010ad600e4dde652294fd669
[]
no_license
cran/sadists
222c55c3cbb6fc42d049d63a76d1d18449bd1df1
7941938a1576fa5cc308c2e5b7d6a0236f03d663
refs/heads/master
2023-09-01T10:47:45.370954
2023-08-21T18:30:02
2023-08-21T19:31:01
31,738,358
0
0
null
null
null
null
UTF-8
R
false
true
3,535
rd
dsumlogchisq.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/sumlogchisq.r \name{sumlogchisq} \alias{sumlogchisq} \alias{dsumlogchisq} \alias{psumlogchisq} \alias{qsumlogchisq} \alias{rsumlogchisq} \title{The sum of the logs of (non-central) chi-squares distribution.} \usage{ dsumlogchisq(x, wts, df, ncp=0, log = FALSE, order.max=6) psumlogchisq(q, wts, df, ncp=0, lower.tail = TRUE, log.p = FALSE, order.max=6) qsumlogchisq(p, wts, df, ncp=0, lower.tail = TRUE, log.p = FALSE, order.max=6) rsumlogchisq(n, wts, df, ncp=0) } \arguments{ \item{x, q}{vector of quantiles.} \item{wts}{the vector of weights. This is recycled against the \code{df, ncp}, but not against the \code{x,q,p,n}.} \item{df}{the vector of degrees of freedom. This is recycled against the \code{wts, ncp}, but not against the \code{x,q,p,n}.} \item{ncp}{the vector of non-centrality parameters. This is recycled against the \code{wts, df}, but not against the \code{x,q,p,n}.} \item{log}{logical; if TRUE, densities \eqn{f} are given as \eqn{\mbox{log}(f)}{log(f)}.} \item{order.max}{the order to use in the approximate density, distribution, and quantile computations, via the Gram-Charlier, Edeworth, or Cornish-Fisher expansion.} \item{p}{vector of probabilities.} \item{n}{number of observations.} \item{log.p}{logical; if TRUE, probabilities p are given as \eqn{\mbox{log}(p)}{log(p)}.} \item{lower.tail}{logical; if TRUE (default), probabilities are \eqn{P[X \le x]}{P[X <= x]}, otherwise, \eqn{P[X > x]}{P[X > x]}.} } \value{ \code{dsumlogchisq} gives the density, \code{psumlogchisq} gives the distribution function, \code{qsumlogchisq} gives the quantile function, and \code{rsumlogchisq} generates random deviates. Invalid arguments will result in return value \code{NaN} with a warning. } \description{ Density, distribution function, quantile function and random generation for the distribution of the weighted sum of logs of non-central chi-squares. } \details{ Let \eqn{X_i \sim \chi^2\left(\delta_i, \nu_i\right)}{X_i ~ chi^2(delta_i, v_i)} be independently distributed non-central chi-squares, where \eqn{\nu_i}{v_i} are the degrees of freedom, and \eqn{\delta_i}{delta_i} are the non-centrality parameters. Let \eqn{w_i} be given constants. Suppose \deqn{Y = \sum_i w_i \log X_i.}{Y = sum w_i log(X_i).} Then \eqn{Y}{Y} follows a weighted sum of log of chi-squares distribution. } \note{ The PDF, CDF, and quantile function are approximated, via the Edgeworth or Cornish Fisher approximations, which may not be terribly accurate in the tails of the distribution. You are warned. The distribution parameters are \emph{not} recycled with respect to the \code{x, p, q} or \code{n} parameters, for, respectively, the density, distribution, quantile and generation functions. This is for simplicity of implementation and performance. It is, however, in contrast to the usual R idiom for dpqr functions. } \examples{ wts <- c(1,-3,4) df <- c(100,20,10) ncp <- c(5,3,1) rvs <- rsumlogchisq(128, wts, df, ncp) dvs <- dsumlogchisq(rvs, wts, df, ncp) qvs <- psumlogchisq(rvs, wts, df, ncp) pvs <- qsumlogchisq(ppoints(length(rvs)), wts, df, ncp) } \references{ Pav, Steven. Moments of the log non-central chi-square distribution. \url{https://arxiv.org/abs/1503.06266} } \seealso{ The product of chi-squares to a power, \code{\link{dprodchisqpow}}, \code{\link{pprodchisqpow}}, \code{\link{qprodchisqpow}}, \code{\link{rprodchisqpow}}. } \author{ Steven E. Pav \email{shabbychef@gmail.com} } \keyword{distribution}
587c9826471bc45db329ccc3afa3b5eb15b11bbe
9ea3df2940ea1796776f00c333a5d8597cf52263
/main.R
93f3ccfb805d7bcc2d92cd453e753ac0c337168f
[]
no_license
enw860/CSC458_Project
5b168a6c0c96092fde5f479e9c35a351b27e4248
4a59647674b4d91657735a4ead1d244faf161019
refs/heads/master
2020-04-05T20:13:34.796370
2019-01-08T16:17:20
2019-01-08T16:17:20
157,170,233
0
0
null
null
null
null
UTF-8
R
false
false
7,480
r
main.R
library(grid) library(gridExtra) library(ggplot2) library(varhandle) library(rlist) ################## begin default settings ##################### # default setting (set your own absoulte path) root_dir <- system("pwd", intern = TRUE) data_dir <- paste(root_dir, "data/", sep="/") img_dir <- paste(root_dir, "plots/", sep="/") # loading sources source(paste(root_dir, "analysis_perflow/analysis_perflow.R", sep="/")) source(paste(root_dir, "analysis_perpack/analysis_perpack.R", sep="/")) source(paste(root_dir, "analysis_RTT/analysis_RTT.R", sep="/")) source(paste(root_dir, "util/util.R", sep="/")) ################## end default settings ##################### ######################### start control box ############ run_perpacket <- bool_switch(1) run_perflow <- bool_switch(1) run_rtt <- bool_switch(1) plot_scatter <- bool_switch(0) plot_ecdf <- bool_switch(0) plot_cdf <- bool_switch(1) ######################### end control box ############## ################# start perpacket analysis ########################## if(run_perpacket){ # Load data print("Loading packets detail") all_packets <- read.table(set_filePath(data_dir, "result.tsv"), sep = "\t" , header = TRUE, stringsAsFactors = FALSE) # cast header length into number all_packets$ip.hdr_len <- suppressWarnings(as.integer(all_packets$ip.hdr_len)) all_packets$ip.len <- suppressWarnings(as.integer(all_packets$ip.len)) # calculate UDP header length udp_header_length <- rep(1, nrow(all_packets)) udp_header_length[is.na(all_packets$udp.length)] <- NA udp_header_dstport <- rep(1, nrow(all_packets)) udp_header_dstport[is.na(all_packets$udp.dstport)] <- NA udp_header_srcport <- rep(1, nrow(all_packets)) udp_header_srcport[is.na(all_packets$udp.srcport)] <- NA udp_header_checksum <- rep(1, nrow(all_packets)) udp_header_checksum[is.na(all_packets$udp.checksum)] <- NA all_packets$udp.hdr_size <- 2 * (udp_header_length + udp_header_dstport + udp_header_srcport + udp_header_checksum) ip_filter <- !is.na(all_packets$ip.len) | (all_packets$ipv6.addr!="") # parsing data into groups IP_packets <- get_IP_packets(all_packets, ip_filter) non_IP_packets <- get_non_IP_packets(all_packets, ip_filter) # plot per-packet analysis bullet point 1 | GENERATE: ./plots/packets_precentage_statistics.png print("Start ploting overall table statistics > GENERATE: ./plots/packets_precentage_statistics.png") plot_precentage_tables(all_packets, img_dir) # plot per-packet analysis bullet point 2 | GENERATE: ./plots/packlen_total_cdf.png & ./plots/packlen_cdfs.png print("Start ploting package length cdf > GENERATE: ./plots/packlen_total_cdf.png & ./plots/packlen_cdfs.png") plot_total_packlen_cdf(all_packets, IP_packets, non_IP_packets, img_dir) # plot per-packet analysis bullet point 2 | GENERATE: ./plots/header_length_cdfs.png print("Start ploting package header length cdf > GENERATE: ./plots/header_length_cdfs.png") plot_total_headerlen_cdf(IP_packets, img_dir) } ################# end perpacket analysis ########################## ################# start perflow analysis ########################## if(run_perflow){ # loading data (~ 3 mins) print("Loading all tcp packets data") tcps <- read.table(set_filePath(data_dir, "tcp_result.tsv"), header=TRUE, sep="\t") tcps <- tcps[!is.na(tcps$tcp.hdr_len), c(1:ncol(tcps))] tcps$ip.hdr_len <- suppressWarnings(as.numeric(unfactor(tcps$ip.hdr_len))) print("Loading all udp packets data") udps <- read.table(set_filePath(data_dir, "udp_result.tsv"), header=TRUE, sep="\t") udps <- udps[!is.na(udps$udp.length), c(1:ncol(udps))] udps$ip.hdr_len <- suppressWarnings(as.numeric(unfactor(udps$ip.hdr_len))) # convert tcp time tcps$frame.time <- convert_time(tcps$frame.time) tcps$frame.time <- tcps$frame.time - tcps$frame.time[1] # convert udp time udps$frame.time <- convert_time(udps$frame.time) udps$frame.time <- udps$frame.time - udps$frame.time[1] # analysis data (~ 8 mins) print("Analysis TCP flows (~ 5 mins)") overall_tcp_flows <- analysis_overall_tcp_flows(tcps) print("Analysis UDP flows (~ 3 mins)") overall_udp_flows <- analysis_overall_udp_flows(udps) # per flow bullet point 1 | GENERATE: ./plots/flows_statistics.png print("Start ploting TCP and UDP flows summary table > GENERATE: ./plots/flows_statistics.png") flows_statistics(img_dir ,overall_tcp_flows, overall_udp_flows) # per flow bullet point 2 | GENERATE: ./plots/durations_cdfs.png print("Start ploting TCP and UDP duration cdf > GENERATE: ./plots/durations_cdfs.png") duration_cdfs(img_dir ,overall_tcp_flows, overall_udp_flows) # per flow bullet point 3 | GENERATE: ./plots/byteSum_cdfs.png print("Start ploting TCP and UDP byte sum cdf > GENERATE: ./plots/byteSum_cdfs.png") byteSum_cdfs(img_dir, overall_tcp_flows, overall_udp_flows) # per flow bullet point 3 | GENERATE: ./plots/packetCount_cdfs.png print("Start ploting TCP and UDP packet count cdf > GENERATE: ./plots/packetCount_cdfs.png ") packetCount_cdfs(img_dir ,overall_tcp_flows, overall_udp_flows) # per flow bullet point 3 | GENERATE: ./plots/overhead_cdfs.png print("Start ploting TCP overhead rate cdf > GENERATE: ./plots/overhead_cdfs.png") overhead_cdfs(img_dir, overall_tcp_flows) # per flow bullet point 4 | GENERATE: ./plots/inter_arriving.png print("Start ploting TCP inter arriving time cdf > GENERATE: ./plots/inter_arriving.png") inter_packet_arrival_cdfs(img_dir ,overall_tcp_flows, overall_udp_flows) # per flow bullet point 5 | GENERATE: ./plots/TCP_exitState_statistics.png print("Start ploting TCP flows exit state table > GENERATE: ./plots/TCP_exitState_statistics.png") exit_state_statistics(img_dir, overall_tcp_flows) } ################# end perflow analysis ########################## ################# start RTT analysis ########################## if(run_rtt){ # get top three TCP connection with largest package count three_longest_packnum_stream <- head(overall_tcp_flows[order(overall_tcp_flows$packets, decreasing = TRUE), 1], 3) print(paste(c("Top three TCP connection with largest package count: ", three_longest_packnum_stream), collapse=" ")) # get top three TCP connection with largest Bytes sum three_largest_bytesum_stream <- head(overall_tcp_flows[order(overall_tcp_flows$total_bytes, decreasing = TRUE), 1], 3) print(paste(c("Top three TCP connection with largest Bytes sum: ", three_largest_bytesum_stream), collapse=" ")) # get top three TCP connection with longest duration three_largest_duration_stream <- head(overall_tcp_flows[order(overall_tcp_flows$duration, decreasing = TRUE), 1], 3) print(paste(c("Top three TCP connection with longest duration: ", three_largest_duration_stream), collapse=" ")) # analysis and draw Sample RTT and estimate RTT analysis_streams <- sort(unique(c(three_longest_packnum_stream, three_largest_bytesum_stream, three_largest_duration_stream))) # plot analysis for all streams in analysis streams print(paste(c("So we need analysis TCP connection: ", analysis_streams), collapse=" ")) analysis_all_stream(img_dir, tcps, analysis_streams) # plot analysis for median streams for top three TCP connections print("Analysis 3 hosts TCP connection over time") analysis_median_host_stream(img_dir, tcps, overall_tcp_flows, 3) } ################# end RTT analysis ##########################
b6a9e142732a56074b4b6f6bf96ddd3b1aee6a4c
6819738573656579a81afe8b292177d5dd81bdbf
/MovieLens_Project.R
9f80ea733828efcb60a120d387fb75a98bbdb819
[]
no_license
rishabhdodeja/MovieLens_Linear
0f73ff46ddee5ea19a99170ad1f3df83bea270f6
62ef5a16a559c53d134c62aa566ef0c1e7959443
refs/heads/master
2022-11-24T14:37:09.009032
2020-07-22T11:45:52
2020-07-22T11:45:52
281,660,577
0
0
null
null
null
null
UTF-8
R
false
false
11,258
r
MovieLens_Project.R
# --- # title: "MovieLens" # author: "Rishabh Singh Dodeja" # date: "July 17, 2020" # --- # This R Script will perform the following task: # 1. Download Movie Lens 10M dataset and required libraries (if needed) # 2. Split the data set into edx(90%) and validation(10%) # 3. Filtering and cleaning the data into final test_set and train_set from validation and edx respectively # 4. Build a Linear Model with predictors : movie mean rating, user effect, genre, and release year # 5. Run Simulations to regularize bias termplot # 6. Calculate RMSE for final Regularized Linear Model # 7. Generate csv file containing entries from validation dataset with following columns: # "userId", "movieId", "rating", "predicted ratings" ################################ # Create edx set, validation set ################################ # Note: this process could take a couple of minutes if(!require(tidyverse)) install.packages("tidyverse", repos = "http://cran.us.r-project.org") if(!require(caret)) install.packages("caret", repos = "http://cran.us.r-project.org") if(!require(data.table)) install.packages("data.table", repos = "http://cran.us.r-project.org") # MovieLens 10M dataset: # https://grouplens.org/datasets/movielens/10m/ # http://files.grouplens.org/datasets/movielens/ml-10m.zip dl <- tempfile() download.file("http://files.grouplens.org/datasets/movielens/ml-10m.zip", dl) ratings <- fread(text = gsub("::", "\t", readLines(unzip(dl, "ml-10M100K/ratings.dat"))), col.names = c("userId", "movieId", "rating", "timestamp")) movies <- str_split_fixed(readLines(unzip(dl, "ml-10M100K/movies.dat")), "\\::", 3) colnames(movies) <- c("movieId", "title", "genres") movies <- as.data.frame(movies) %>% mutate(movieId = as.numeric(levels(movieId))[movieId], title = as.character(title), genres = as.character(genres)) movielens <- left_join(ratings, movies, by = "movieId") # Validation set will be 10% of MovieLens data set.seed(1, sample.kind="Rounding") # if using R 3.5 or earlier, use `set.seed(1)` instead test_index <- createDataPartition(y = movielens$rating, times = 1, p = 0.1, list = FALSE) edx <- movielens[-test_index,] temp <- movielens[test_index,] # Make sure userId and movieId in validation set are also in edx set validation <- temp %>% semi_join(edx, by = "movieId") %>% semi_join(edx, by = "userId") # Add rows removed from validation set back into edx set removed <- anti_join(temp, validation) edx <- rbind(edx, removed) rm(ratings, movies, test_index, temp, movielens, removed) ################################ # Data Filtering and cleaning ################################ # Note: this process removes features/colums that are not required by our Predictor Model # we will only be taking userId, genre, and release year into account. # This selection is based on insights gained from data exploration, explained in Rmd and report. # We will make train set from edx set, our model will be trained on this set train_set <- edx %>% select(userId, movieId, rating, title,genres) %>% mutate(year = as.numeric(str_sub(title,-5,-2))) # We will make test_set for validation set, our model will be testes/validated on this set test_set <- validation %>% select(userId, movieId, rating, title,genres) %>% mutate(year = as.numeric(str_sub(title,-5,-2))) # remove rest of the objects/datasets to free up disk space rm(edx,validation) ################################ # Model Evaluation ################################ # Define Mean Absolute Error (MAE) MAE <- function(true_ratings, predicted_ratings){ mean(abs(true_ratings - predicted_ratings)) } # Define Mean Squared Error (MSE) MSE <- function(true_ratings, predicted_ratings){ mean((true_ratings - predicted_ratings)^2) } # Define Root Mean Squared Error (RMSE) RMSE <- function(true_ratings, predicted_ratings){ sqrt(mean((true_ratings - predicted_ratings)^2)) } # Create Results Table result <- tibble(Method = "Project Goal", RMSE = 0.8649, MSE = NA, MAE = NA); ################################ # Modelling and Results ################################ #LINEAR MODEL# #Calculating Movie Mean Ratings print("evaluating Movie-Mean-Model...") #calculating movie_mean from train_set movie_mean = train_set %>% group_by(movieId) %>% summarise(mu=mean(rating)) #Join Predicted values with test_set y_hat_mu = test_set %>% left_join(movie_mean,by="movieId") %>% .$mu # Update the Results table result <- bind_rows(result, tibble(Method = "Movie Mean", RMSE = RMSE(test_set$rating, y_hat_mu), MSE = MSE(test_set$rating, y_hat_mu), MAE = MAE(test_set$rating, y_hat_mu))); print(result) #Calculating User Effect Bias terms print("Calculating User-Effect and evaluating new model...") user_effect = train_set %>% left_join(movie_mean,by="movieId") %>% group_by(userId) %>% summarise(bu = mean(rating-mu)); # Join predicted values with test_set y_hat_bu = test_set %>% left_join(movie_mean, by="movieId") %>% left_join(user_effect,by="userId") %>% mutate(pred=mu+bu) %>% .$pred; result <- bind_rows(result, tibble(Method = "Movie Mean + bu", RMSE = RMSE(test_set$rating, y_hat_bu), MSE = MSE(test_set$rating, y_hat_bu), MAE = MAE(test_set$rating, y_hat_bu))) # Show the RMSE improvement print(result) #Calculating Genre-Effect bias terms print ("Calculating Genre-effect and including in evaluation...") #calculating genre effect from train_set genre_effect = train_set %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>% group_by(genres) %>% summarise(bg = mean(rating-mu-bu)); #Joining predicted values with test_set y_hat_bg = test_set %>% left_join(movie_mean, by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% mutate(pred=mu+bu +bg) %>% .$pred; result <- bind_rows(result, tibble(Method = "Movie Mean + bu + bg", RMSE = RMSE(test_set$rating, y_hat_bg), MSE = MSE(test_set$rating, y_hat_bg), MAE = MAE(test_set$rating, y_hat_bg))) # Show the RMSE improvement print(result) #Calculating Release Year Effect bias terms print("calclating Release-Year Effect and evaluating..."); #calculating year effect from train_set year_effect = train_set %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>%left_join(genre_effect,by="genres") %>% group_by(year) %>% summarise(by = mean(rating-mu-bu-bg)); # Join predictions with test_set y_hat_by = test_set %>% left_join(movie_mean, by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% left_join(year_effect,be="year")%>% mutate(pred=mu+bu +bg+by) %>% .$pred; #update results table result = bind_rows(result, tibble(Method = "Movie Mean + bu + bg + by", RMSE = RMSE(test_set$rating, y_hat_by), MSE = MSE(test_set$rating, y_hat_by), MAE = MAE(test_set$rating, y_hat_by))); # Show the RMSE improvement print(result) #REGULARIZATION# #defining regularization fuction regularization <- function(lambda, trainset, testset){ # Movie Mean movie_mean = trainset %>% group_by(movieId) %>% summarise(mu=mean(rating)); # User effect (bu) user_effect = trainset %>% left_join(movie_mean,by="movieId") %>% group_by(userId) %>% summarise(bu = sum(rating-mu)/(n()+lambda)); #Genre effect (bg) genre_effect = trainset %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>% group_by(genres) %>% summarise(bg = sum(rating-mu-bu)/(n()+lambda)); #Year effect (by) year_effect = trainset %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% group_by(year) %>% summarise(by = sum(rating-mu-bu-bg)/(n()+lambda)); # Prediction: mu + bu + bg + by predicted_ratings = testset %>% left_join(movie_mean, by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% left_join(year_effect,be="year")%>% mutate(pred=mu+bu +bg+by) %>% .$pred; return(RMSE(testset$rating,predicted_ratings)); } #Running Regularization Simulation to get optimal value of lambda print("Now running Regularization to tune lambda...") # Define a set of lambdas to tune lambdas = seq(0, 10, 0.25) # Tune lambda rmses = sapply(lambdas, regularization, trainset = train_set, testset = test_set) # Plot the lambda vs RMSE tibble(Lambda = lambdas, RMSE = rmses) %>% ggplot(aes(x = Lambda, y = RMSE)) + geom_point() + ggtitle("Regularization") #picking lambda with lowest RMSE lambda = lambdas[which.min(rmses)]; #Evaluating ofr optimal lambda print("Regularization done. Evaluating Final Model...") # Movie Mean movie_mean = train_set %>% group_by(movieId) %>% summarise(mu=mean(rating)); # User effect (bu) user_effect = train_set %>% left_join(movie_mean,by="movieId") %>% group_by(userId) %>% summarise(bu = sum(rating-mu)/(n()+lambda)); #Genre effect (bg) genre_effect = train_set %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>% group_by(genres) %>% summarise(bg = sum(rating-mu-bu)/(n()+lambda)); #Year effect (by) year_effect = train_set %>% left_join(movie_mean,by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% group_by(year) %>% summarise(by = sum(rating-mu-bu-bg)/(n()+lambda)); # Prediction: mu + bu + bg + by y_hat_reg = test_set %>% left_join(movie_mean, by="movieId") %>% left_join(user_effect,by="userId") %>% left_join(genre_effect,by="genres") %>% left_join(year_effect,be="year")%>% mutate(pred=mu+bu +bg+by) %>% .$pred; # Final Result Results Table result <- bind_rows(result, tibble(Method = "Regularized (mu + bu + bg + by)", RMSE = RMSE(test_set$rating, y_hat_reg), MSE = MSE(test_set$rating, y_hat_reg), MAE = MAE(test_set$rating, y_hat_reg))); # Display Final RMSE table print(result) print(paste("RMSE for Final Model is:", RMSE(test_set$rating, y_hat_reg))) ################################ # Save Predictions ################################ predictions = test_set %>% select(movieId,userId,rating) %>% mutate(predictedRatings = y_hat_reg) write.csv(predictions, "predictions.csv",row.names=FALSE)
3018adde5f6371f7e1f7e0ebb653f0a4719ebb16
6ae01e2f3d6134c2b008cca6a174951414ceaa05
/cachematrix.R
0634437a9c4ae01e340d9a468ac3bd182a73aa75
[]
no_license
markcovello-home/ProgrammingAssignment2
ae03f877ce10adba2e1e13fbf485eb372ec3f0ee
2403a8d181755c624d8ed429db39bf238986508b
refs/heads/master
2020-12-26T02:33:07.934878
2015-04-25T05:49:46
2015-04-25T05:49:46
34,495,827
0
0
null
2015-04-24T03:26:52
2015-04-24T03:26:51
null
UTF-8
R
false
false
2,582
r
cachematrix.R
## {assignment root)/source/cachematrix.R code to read/write a matrix from cache ## Not sure I get most of this, but what is sure is we must: ## 1. Test to see if the matrix is in the solve cache ## 2. if matrix is not in cache, calculate inverse and cache it ## 3. If the matrix is in cache then just return it ## two functions accomplish this makeCacheMatrix() makes a, "special," matrix ## that chacheSolve() can use to either calculate or extract from cache ## the inverse of the matrix that was the argument originally passed to makeCacheMatrix() ## makeCacheMatrix() takes an invertible matrix and returns a, "special matrix," ## (in fact it's a list and it looks a lot like a list of instructions) ## that chacheSolve() can use to return the inverse of the matrix makeCacheMatrix <- function(x = matrix()) { ## initialize the value of the inverse to NULL, ## matrixinverse is the token to identify the inverse (solution) in this function matrixinverse <- NULL ## This simple function just puts it's argument into the cache(parent env) ## th token that holds the cached object in the parent env is x set <- function(y) { x <<- y ## matrixinverse <<- NULL } ## gets the value of the inverse; x is the token foe the cached inverse get <- function() x #calculates the inverse of non-singular matrix via the solve function setinverse <- function(solve) matrixinverse <<- solve # gets the inverse getinverse <- function() matrixinverse ## passes the value of the function makeCacheMatrix list(set = set, get = get, setinverse = setinverse, getinverse = getinverse) } ## chacheSolve() takes the special matrix from makeCacheMatrix() and returns either the ## previously calculated cached inverse of the original input matrix ## or calculates its inverse directly cacheSolve<- function(x, ...) { ## load the cache into matrixinverse matrixinverse <- x$getinverse() #if the cache is not null just return it and we're done if(!is.null(matrixinverse)) { message("getting cached data - Inverse of the matrix") return(matrixinverse) } #otherwise , calculate the inverse and send it back with a message saying it was calculated. data <- x$get() message("not cached - Calculating Inverse") matrixinverse <- solve(data, ...) x$setinverse(matrixinverse) matrixinverse }
66acf7a5ecb18fc01899115b22e379a0763b54d0
d1e2a6d642015a3eb37162808648ff32bfaa0bce
/data/processed/process_train_val_history.R
fa81c3e83495942cc93056e58cc9e826037762bb
[]
no_license
TeamMacLean/effector_class_ms
a889cbab122a9d8fe0566cbabd414da197104ec7
34f3f71ebf64e0e4863fe2e4b809e2dc798be7c8
refs/heads/master
2023-06-04T16:06:46.205768
2020-06-17T08:35:49
2020-06-17T08:35:49
261,469,416
0
0
null
null
null
null
UTF-8
R
false
false
4,675
r
process_train_val_history.R
# Load raw data ---- bacteria_cnn_lstm_train <- data.table::fread("data/raw/bacteria_hist_cnn_lstm_train.csv") bacteria_cnn_lstm_val <- data.table::fread("data/raw/bacteria_hist_cnn_lstm_val.csv") bacteria_cnn_gru_train <- data.table::fread("data/raw/bacteria_hist_cnn_gru_train.csv") bacteria_cnn_gru_val <- data.table::fread("data/raw/bacteria_hist_cnn_gru_val.csv") bacteria_lstm_emb_train <- data.table::fread("data/raw/bacteria_hist_lstm_emb_train.csv") bacteria_lstm_emb_val <- data.table::fread("data/raw/bacteria_hist_lstm_emb_val.csv") bacteria_gru_emb_train <- data.table::fread("data/raw/bacteria_hist_gru_emb_train.csv") bacteria_gru_emb_val <- data.table::fread("data/raw/bacteria_hist_gru_emb_val.csv") fungi_cnn_lstm_train <- data.table::fread("data/raw/fungi_hist_cnn_lstm_train.csv") fungi_cnn_lstm_val <- data.table::fread("data/raw/fungi_hist_cnn_lstm_val.csv") fungi_cnn_gru_train <- data.table::fread("data/raw/fungi_hist_cnn_gru_train.csv") fungi_cnn_gru_val <- data.table::fread("data/raw/fungi_hist_cnn_gru_val.csv") fungi_lstm_emb_train <- data.table::fread("data/raw/fungi_hist_lstm_emb_train.csv") fungi_lstm_emb_val <- data.table::fread("data/raw/fungi_hist_lstm_emb_val.csv") fungi_gru_emb_train <- data.table::fread("data/raw/fungi_hist_gru_emb_train.csv") fungi_gru_emb_val <- data.table::fread("data/raw/fungi_hist_gru_emb_val.csv") oomycete_cnn_lstm_train <- data.table::fread("data/raw/oomycete_hist_cnn_lstm_train.csv") oomycete_cnn_lstm_val <- data.table::fread("data/raw/oomycete_hist_cnn_lstm_val.csv") oomycete_cnn_gru_train <- data.table::fread("data/raw/oomycete_hist_cnn_gru_train.csv") oomycete_cnn_gru_val <- data.table::fread("data/raw/oomycete_hist_cnn_gru_val.csv") oomycete_lstm_emb_train <- data.table::fread("data/raw/oomycete_hist_lstm_emb_train.csv") oomycete_lstm_emb_val <- data.table::fread("data/raw/oomycete_hist_lstm_emb_val.csv") oomycete_gru_emb_train <- data.table::fread("data/raw/oomycete_hist_gru_emb_train.csv") oomycete_gru_emb_val <- data.table::fread("data/raw/oomycete_hist_gru_emb_val.csv") # Bind results together ---- bacteria_train_data <- dplyr::bind_rows( bacteria_cnn_lstm_train %>% mutate(model = "CNN-LSTM"), bacteria_cnn_gru_train %>% mutate(model = "CNN-GRU"), bacteria_lstm_emb_train %>% mutate(model = "LSTM-Emb"), bacteria_gru_emb_train %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) bacteria_val_data <- dplyr::bind_rows( bacteria_cnn_lstm_val %>% mutate(model = "CNN-LSTM"), bacteria_cnn_gru_val %>% mutate(model = "CNN-GRU"), bacteria_lstm_emb_val %>% mutate(model = "LSTM-Emb"), bacteria_gru_emb_val %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) fungi_train_data <- dplyr::bind_rows( fungi_cnn_lstm_train %>% mutate(model = "CNN-LSTM"), fungi_cnn_gru_train %>% mutate(model = "CNN-GRU"), fungi_lstm_emb_train %>% mutate(model = "LSTM-Emb"), fungi_gru_emb_train %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) fungi_val_data <- dplyr::bind_rows( fungi_cnn_lstm_val %>% mutate(model = "CNN-LSTM"), fungi_cnn_gru_val %>% mutate(model = "CNN-GRU"), fungi_lstm_emb_val %>% mutate(model = "LSTM-Emb"), fungi_gru_emb_val %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) oomycete_train_data <- dplyr::bind_rows( oomycete_cnn_lstm_train %>% mutate(model = "CNN-LSTM"), oomycete_cnn_gru_train %>% mutate(model = "CNN-GRU"), oomycete_lstm_emb_train %>% mutate(model = "LSTM-Emb"), oomycete_gru_emb_train %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) oomycete_val_data <- dplyr::bind_rows( oomycete_cnn_lstm_val %>% mutate(model = "CNN-LSTM"), oomycete_cnn_gru_val %>% mutate(model = "CNN-GRU"), oomycete_lstm_emb_val %>% mutate(model = "LSTM-Emb"), oomycete_gru_emb_val %>% mutate(model = "GRU-Emb") ) %>% dplyr::mutate(epochs = V1 + 1) %>% dplyr::select(model, epochs, acc, loss) # Save data ---- bacteria_train_data %>% data.table::fwrite("data/processed/bacteria_hist_train.csv") bacteria_val_data %>% data.table::fwrite("data/processed/bacteria_hist_val.csv") fungi_train_data %>% data.table::fwrite("data/processed/fungi_hist_train.csv") fungi_val_data %>% data.table::fwrite("data/processed/fungi_hist_val.csv") oomycete_train_data %>% data.table::fwrite("data/processed/oomycete_hist_train.csv") oomycete_val_data %>% data.table::fwrite("data/processed/oomycete_hist_val.csv")
ec5f19cada22685cd9475e887e4197e32f605d5e
94bd617b377a58ebd0fbbd0d70a0a43362a1108d
/R/seed_generators.R
0ce3e3aa6cd74fce1ba66efd8acc61fdd4ba7f77
[ "MIT" ]
permissive
Maschette/jasmines
36585e07b95676a66c05154ee3e753e37116ba77
1218da72c52ca7c93d2a0a72bf03fd715a6fc294
refs/heads/master
2020-11-24T10:38:04.772951
2019-12-16T01:03:50
2019-12-16T01:03:50
228,111,310
0
0
null
2019-12-15T01:06:02
2019-12-15T01:06:01
null
UTF-8
R
false
false
3,868
r
seed_generators.R
#' Seed shaped like a teardrop #' #' @param n number of dot points #' @param m exponent #' #' @return tibble #' @export seed_teardrop <- function(n = 100, m = 3) { t <- seq(0, 2*pi, length.out = n) tibble::tibble( x = sin(t) * (sin(t/2))^m, y = cos(t), id = 1 ) } #' Seed shaped like a heart #' #' @param n number of dot points #' #' @return tibble #' @export seed_heart <- function(n = 100) { t <- seq(0, 2*pi, length.out = n) tibble::tibble( x = (16*sin(t)^3) / 17, y = (13*cos(t) - 5*cos(2*t) - 2*cos(3*t) - cos(4*t)) / 17, id = 1 ) } #' Seed shaped like concentric rings #' #' @param points Total number of interior points #' @param rings How many rings to spread across? #' @param size Radius of the outermost ring #' @export seed_disc <- function(points = 1000, rings = 10, size = 1) { radius <- size * (1:rings)/rings circumference <- 2 * pi * radius proportion <- circumference / sum(circumference) counts <- round(points * proportion) unfold <- function(radius, grain, id) { theta <- seq(0, 2*pi, length.out = grain + 1) theta <- theta[-1] return(tibble::tibble( x = radius * cos(theta), y = radius * sin(theta), id = id )) } seed <- purrr::pmap_dfr( .l = list(radius, counts, 1:length(radius)), .f = unfold ) return(seed) } #' Seed shaped from text #' #' @param text the message #' #' @return a tibble #' @export seed_text <- function(text) { char_map <- make_dotted_letters() char_map <- dplyr::filter(char_map, value == 1) char_set <- stringr::str_split(text, "", simplify = TRUE) dots <- purrr::map_dfr( .x = char_set, .f = ~ dplyr::filter(char_map, char == .x), .id = "char_ind" ) %>% dplyr::mutate( char_ind = as.numeric(char_ind), x = x + 6 * char_ind, id = 1:nrow(.) ) %>% dplyr::mutate( # to match scale used in bridges x = x/10, y = y/10 ) %>% dplyr::select(char, char_ind, x, y, id) return(dots) } #' Seed from a set of random sticks #' #' @param n how many sticks #' @param grain how many points along each stick #' #' @return a tibble with columns x, y and id #' @export seed_sticks <- function(n = 10, grain = 1000) { make_stick <- function(id, grain) { return(tibble::tibble( x = seq(stats::runif(1), stats::runif(1), length.out = grain), y = seq(stats::runif(1), stats::runif(1), length.out = grain), id = id )) } points <- purrr::map_dfr(1:n, make_stick, grain = grain) return(points) } #' Seed with evenly spaced rows #' #' @param n how many rows #' @param grain how many points along each row #' @param vertical flip the x/y co-ords to produce columns? #' #' @return a tibble with columns x, y and id #' @export seed_rows <- function(n = 10, grain = 1000, vertical = FALSE) { make_row <- function(id, grain, vertical = FALSE) { if(!vertical) { return(tibble::tibble( x = seq(0, 1, length.out = grain), y = id/(n+1), id = id )) } else{ return(tibble::tibble( x = id/(n+1), y = seq(1, 0, length.out = grain), id = id )) } } points <- purrr::map_dfr(1:n, make_row, grain = grain, vertical = vertical) return(points) } #' Seed with a random set of bubbles #' #' @param n how many bubbles #' @param grain how many points along each stick #' #' @return a tibble with columns x, y and id #' @export seed_bubbles <- function(n = 2, grain = 1000) { make_bubble <- function(id, grain) { radius <- stats::runif(1) origin_x <- stats::runif(1) origin_y <- stats::runif(1) th <- seq(0, 2*pi, length.out = grain) return(tibble::tibble( x = radius * cos(th) + origin_x, y = radius * sin(th) + origin_y, id = id )) } points <- purrr::map_dfr(1:n, make_bubble, grain = grain) return(points) }
924220bde9e48d6b8d037eed2a0862da81d11a7d
1da4a08a9e5a33ae868187de57c6ac324b627fce
/03.phylogenetic-stats.r
43fd902269d678c07483833d28ec99bf804af98b
[]
no_license
basil-yakimov/meadow-structure
a211e35401661e6983cae69bd125f6bc3ca863b3
eaa420636b5acdd0c081a1c2f6da55dc1a3a1ed7
refs/heads/master
2020-07-22T19:09:26.136503
2019-11-15T04:50:24
2019-11-15T04:50:24
207,300,289
0
0
null
null
null
null
UTF-8
R
false
false
6,137
r
03.phylogenetic-stats.r
load("clean.data/meadows-2018.rda") load("clean.data/meadows-phylo-ses-ao.rda") meta <- meta2 plot.year <- function(y , ind ) { dd <- meta[[ind]][meta$year == y] ss <- factor(meta$site[meta$year == y], levels = paste0(c("a", "f1", "f2", "m"), "-", substr(y, 3, 4))) boxplot(dd ~ ss, col = c("tomato", "steelblue", "skyblue", "limegreen"), ylab = ind) } plot.year(2014, "mpd.z") op <- par(mfrow = c(2, 5), mar = c(2, 4, 0.5, 0.5)) plot.year(2014, "mpd.z") plot.year(2015, "mpd.z") plot.year(2016, "mpd.z") plot.year(2017, "mpd.z") plot.year(2018, "mpd.z") plot.year(2014, "mpd.a.z") plot.year(2015, "mpd.a.z") plot.year(2016, "mpd.a.z") plot.year(2017, "mpd.a.z") plot.year(2018, "mpd.a.z") par(op) #---# df <- data.frame(mpd.z = meta$mpd.z, mpd.a.z = meta$mpd.a.z, mntd.z = meta$mntd.z, mntd.a.z = meta$mntd.a.z, year = meta$year, site = sapply(strsplit(as.character(meta$site), "-"), function(x) x[[1]]), id = paste0(sapply(strsplit(as.character(meta$site), "-"), function(x) x[[1]]), "-", c(rep(1:50, 2), rep(1:65, 15)))) df$numsite <- as.numeric(df$site) dfa <- data.frame(mpd.z = tapply(df$mpd.z, df$id, mean), mpd.a.z = unname(tapply(df$mpd.a.z, df$id, mean)), mntd.z = unname(tapply(df$mntd.z, df$id, mean)), mntd.a.z = unname(tapply(df$mntd.a.z, df$id, mean))) dfa$site <- sapply(strsplit(as.character(rownames(dfa)), "-"), function(x) x[[1]]) dfa$site <- factor(dfa$site) #---------------------------------------------------------# fit <- lm(mpd.z ~ year + id, data = df) ( an <- anova(fit) ) eta <- an[, 2] / sum(an[, 2]) names(eta) <- row.names(an) eta * 100 fit <- lm(mpd.a.z ~ year + id, data = df) ( an <- anova(fit) ) eta <- an[, 2] / sum(an[, 2]) names(eta) <- row.names(an) eta * 100 fit <- lm(mntd.z ~ year + id, data = df) ( an <- anova(fit) ) eta <- an[, 2] / sum(an[, 2]) names(eta) <- row.names(an) eta * 100 fit <- lm(mntd.a.z ~ year + id, data = df) ( an <- anova(fit) ) eta <- an[, 2] / sum(an[, 2]) names(eta) <- row.names(an) eta * 100 #---------------------------------------------------------# col4 <- c("tomato", "steelblue", "skyblue", "limegreen") op <- par(mfcol = c(2, 2)) boxplot(mpd.z ~ site, dfa, col = col4, ylab = "SES MPD") boxplot(mpd.a.z ~ site, dfa, col = col4, ylab = expression("SES "*MPD[a])) boxplot(mntd.z ~ site, dfa, col = col4, ylab = "SES MNTD") boxplot(mntd.a.z ~ site, dfa, col = col4, ylab = expression("SES "*MNTD[a])) par(op) #---------------------------------------------------------# library(multcompView) library(reshape) cld <- function(res) { require(multcompView) require(reshape) a <- melt(res$p.value) a.cc <- na.omit(a) a.pvals <- a.cc[, 3] names(a.pvals) <- paste(a.cc[, 2], a.cc[, 1], sep="-") multcompLetters(a.pvals)$Letters } fit <- lm(mpd.z ~ site, dfa) anova(fit) res <- pairwise.t.test(dfa$mpd.z, dfa$site) abc <- cld(res) pp <- tapply(dfa$mpd.z, dfa$site, function(x) t.test(x)$p.value) abc[pp < 0.05] <- paste0(abc[pp < 0.05], "(*)") boxplot(mpd.z ~ site, dfa, col = col4, ylab = "SES MPD", ylim = c(-.5, .5), axes = F, cex.lab = 1) axis(1, at = 1:4, cex.axis = 1, labels = c("поле", "залежь 1", "залежь 2", "луг")) axis(2, cex.axis = 1) box() text(x = 1:4, y = 0.45, labels = abc, cex = 1) #---# fit <- lm(mpd.a.z ~ site, dfa) anova(fit) res <- pairwise.t.test(dfa$mpd.a.z, dfa$site) abc <- cld(res) pp <- tapply(dfa$mpd.a.z, dfa$site, function(x) t.test(x)$p.value) abc[pp < 0.05] <- paste0(abc[pp < 0.05], "(*)") boxplot(mpd.a.z ~ site, dfa, col = col4, ylab = expression("SES "*MPD[a]), ylim = c(-.5, .5), axes = F, cex.lab = 1) axis(1, at = 1:4, cex.axis = 1, labels = c("поле", "залежь 1", "залежь 2", "луг")) axis(2, cex.axis = 1) box() text(x = 1:4, y = 0.45, labels = abc, cex = 1) #________________________________________________________________________________________ boxplotm <- function(x, y, i){ res <- pairwise.t.test(x, y) abc <- cld(res) pp <- tapply(x, y, function(x) t.test(x)$p.value) abc[pp < 0.05] <- paste0(abc[pp < 0.05], "(*)") col4 <- c("tomato", "steelblue", "skyblue", "limegreen") ylab <- c("SES MPD", expression("SES "*MPD[a]), "SES MNTD", expression("SES "*MNTD[a])) boxplot(x ~ y, col = col4, ylab = ylab[i], ylim = c(min(x), max(x)+0.21), axes = F, cex.lab = 1) axis(1, at = 1:4, cex.axis = 1, labels = c("поле", "залежь 1", "залежь 2", "луг")) axis(2, cex.axis = 1) box() text(x = 1:4, y = max(x) + 0.15, labels = abc, cex = 1) } png("figures/ses-phylo-angiosperm-only.png", width = 1000, height = 1000) op <- par(mfcol = c(2,2), mar = c(4, 4.1, 1, 1), cex = 1.5) boxplotm(dfa$mpd.z, dfa$site, 1) boxplotm(dfa$mpd.a.z, dfa$site, 2) boxplotm(dfa$mntd.z, dfa$site, 3) boxplotm(dfa$mntd.a.z, dfa$site, 4) par(op) dev.off() #________________________________________________________________________________________ load("clean.data/meadows-phylo-ses.rda") df <- data.frame(mpd.z = meta$mpd.z, mpd.a.z = meta$mpd.a.z, mntd.z = meta$mntd.z, mntd.a.z = meta$mntd.a.z, year = meta$year, site = sapply(strsplit(as.character(meta$site), "-"), function(x) x[[1]]), id = paste0(sapply(strsplit(as.character(meta$site), "-"), function(x) x[[1]]), "-", c(rep(1:50, 2), rep(1:65, 15)))) df$numsite <- as.numeric(df$site) dfa <- data.frame(mpd.z = tapply(df$mpd.z, df$id, mean), mpd.a.z = unname(tapply(df$mpd.a.z, df$id, mean)), mntd.z = unname(tapply(df$mntd.z, df$id, mean)), mntd.a.z = unname(tapply(df$mntd.a.z, df$id, mean))) dfa$site <- sapply(strsplit(as.character(rownames(dfa)), "-"), function(x) x[[1]]) dfa$site <- factor(dfa$site) png("figures/ses-phylo-all-species.png", width = 1000, height = 1000) op <- par(mfcol = c(2,2), mar = c(4, 4.1, 1, 1), cex = 1.5) boxplotm(dfa$mpd.z, dfa$site, 1) boxplotm(dfa$mpd.a.z, dfa$site, 2) boxplotm(dfa$mntd.z, dfa$site, 3) boxplotm(dfa$mntd.a.z, dfa$site, 4) par(op) dev.off()
fb4fa36f17d8ccf6ab778504403df54d64745e0d
524b38b3eba37f9ccfb169b7fdb92286a5bbfd36
/tests/testthat/testcorrmatrix.R
6d96b52585a66138dbb9086b069ec911ecc29089
[]
no_license
lago1970/jmv
b623a4bad1191cdf864c6c3a3ab0c450eaaa3b68
41247ccd0a9ea985f218494b1fc810e669ec6ecb
refs/heads/master
2020-03-08T05:57:33.352626
2018-04-02T01:35:47
2018-04-02T01:39:46
null
0
0
null
null
null
null
UTF-8
R
false
false
481
r
testcorrmatrix.R
context('corrmatrix') test_that('corrmatrix works', { x <- as.factor(c(NA,rep(c("a", "b"), 6))) y <- c(8,51,2,74,1,91,5,25,1,59,5,32,7) # breaks equality of variance z <- c(2,NA,NaN,3,-1,-2,1,1,-2,2,-2,-3,3) w <- c(0,4,19,5,9,15,1,4,19,10,13,7,5) data <- data.frame(x = x, y = y, z = z, w = w) # expect_error(jmv::corrMatrix(data, c("x","y","z","w")), "'x' must be a numeric vector", fixed=TRUE) corrmatrix <- jmv::corrMatrix(data, c("y","w")) })
309b95ba135a3122aa693e0335301ee53b268548
cba10b84d2cc708dd66148a4511451d77a92a7c5
/man/selShapes.Rd
48fa73a54a538716679125bb54c82deef3dc6240
[ "LicenseRef-scancode-warranty-disclaimer" ]
no_license
r4ss/r4ss
03e626ae535ab959ff8109a1de37e3e8b44fe7ad
0ef80c1a57e4a05e6172338ddcb0cda49530fa93
refs/heads/main
2023-08-17T08:36:58.041402
2023-08-15T21:42:05
2023-08-15T21:42:05
19,840,143
35
57
null
2023-07-24T20:28:49
2014-05-16T00:51:48
R
UTF-8
R
false
true
430
rd
selShapes.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/selShapes.R \name{selShapes} \alias{selShapes} \title{Launch a shiny app that displays various selectivity curves} \usage{ selShapes() } \description{ This app is hosted at https://connect.fisheries.noaa.gov/ss3-helper/ } \author{ Allan C. Hicks, Andrea M. Havron, Ian G. Taylor, Kathryn L. Doering inspired by tcl/tk code written by Tommy Garrison }
d761806ac566c82816c1b264eedee845728b14e1
2fb65d442efadbc3a1db41fcf25fed8958c4e04f
/man/readable.files.Rd
04998efee8bcdfb96c5e9f9c38ad9947ffe0160d
[ "MIT" ]
permissive
dfsp-spirit/freesurferformats
8f507d8b82aff7c34b12e9182893007064e373b9
6cf9572f46608b7bb53887edd10dfed10e16e13d
refs/heads/master
2023-07-25T00:28:09.021237
2023-07-19T07:29:07
2023-07-19T07:29:07
203,574,524
22
3
NOASSERTION
2023-07-19T07:29:09
2019-08-21T11:57:16
R
UTF-8
R
false
true
919
rd
readable.files.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/helpers.R \name{readable.files} \alias{readable.files} \title{Find files with the given base name and extensions that exist.} \usage{ readable.files( filepath, precedence = c(".mgh", ".mgz"), error_if_none = TRUE, return_all = FALSE ) } \arguments{ \item{filepath}{character string, path to a file without extension} \item{precedence}{vector of character strings, the file extensions to check. Must include the dot (if you expect one).} \item{error_if_none}{logical, whether to raise an error if none of the files exist} \item{return_all}{logical, whether to return all readable files instead of just the first one} } \value{ character string, the path to the first existing file (or `NULL` if none of them exists). } \description{ Note that in the current implementation, the case of the filepath and the extension must match. }
893b0b5266100f3bfbef330bf8bc7ec4b720a6af
07a57621be185d493c6026cdfb972a5cf0376c09
/DecompMex/R/9_Infection_notHom.R
ef05e9872def399dfe8e7e539fb545670b7517b6
[]
no_license
timriffe/DecompMex
fe47b21c8b6796eeb3f6f3b8e87bddc02d617d1c
b92b5aa927cc52d6c1be6b741eaa7bf1ebdd208f
refs/heads/master
2020-04-07T03:59:40.373849
2018-08-07T14:05:23
2018-08-07T14:05:23
42,098,790
0
1
null
2018-08-07T13:56:13
2015-09-08T08:07:42
TeX
UTF-8
R
false
false
882
r
9_Infection_notHom.R
# test of female homicide spike 2009 head(data) # g7 is homicide load('Data/Counts&Rates_1990-2015Mex.RData') head(Data_rates) dim(Data_Counts) dim(data) data <- data[with(data, order(state,year,sex,age)), ] Data_Counts <- Data_Counts[with(Data_Counts, order(state,year,sex,age)), ] Data_rates <- Data_rates[with(Data_rates, order(state,year,sex,age)), ] Data_rates$Exp <- Data_rates$Pop Data_rates[,standage := sum(Exp), by=list(year,sex,age)] strate <- function(rate,standardage,age,from=0,to=14){ keep <- age >= from & age <= to sum(rate[keep] * standardage[keep]) / sum(standardage[keep]) } setnames(Data_rates,as.character(1:12),letters[1:12]) asdr <- Data_rates[ ,list(asdr = strate(a,standage,age,from=50,to=59)),by=list(state,year,sex)] mat <- acast(asdr[sex == 1, ], state ~ year, value.var = "asdr") matplot(1990:2015, t(mat), type = 'l', lty = 1) abline(v=2009)
5d21301d8e44aa3629aed550ca794ca43ad2927c
e244706f842fd980ad5887c953817d63f0a329dc
/gapminder_practice.R
c1aabf455bda018e3e838efdaee8b3d53de39609
[]
no_license
cavfan96/tidyverse_with_gapminder
15ce1c5be1047aecd5c4b27219dffd83b63bba2e
5c915703fd1903e8678528c8b7c1536d0a79a9aa
refs/heads/master
2022-12-07T21:13:37.976236
2020-08-23T17:19:43
2020-08-23T17:19:43
289,731,013
0
0
null
null
null
null
UTF-8
R
false
false
5,905
r
gapminder_practice.R
# tidyverse library(gapminder) library(dplyr) gapminder ## Filter Verb # filters show a subset of the data # filter for just the year 2007 using the pipe %>% # piping simply is creating a df on the fly without # changing your original data frame unless you assign it # as a new dataframe gapminder %>% filter(year == 2007) # filter for the US - need quotes around text gapminder %>% filter(country == "United States") # filter by two conditions gapminder %>% filter(country == "United States", year == 2007) ## Arrange Verb # Sorts the data gapminder %>% arrange(gdpPercap) gapminder %>% arrange(desc(gdpPercap)) # combining verbs to find richest countries in 2007 descending order gapminder %>% filter(year == 2007) %>% arrange(desc(gdpPercap)) ## Mutate Verb # changes the data value in a new value, doesn't effect the original data gapminder %>% mutate(pop = pop / 1000000) # add a variable gapminder %>% mutate(gdp = gdpPercap * pop) %>% arrange(desc(gdp)) # Create, filter, and sort the data by combining verbs gapminder %>% mutate(gdp = gdpPercap * pop) %>% filter(year == 2007) %>% arrange(desc(gdp)) ## Data Visualization ## Data, Aesthetic, Layer library(ggplot2) # create a new dataframe for 2007 gapminder_2007 <- gapminder %>% filter(year == 2007) gapminder_2007 # ggplots start with the df and then an aesthetic # then you add a layer (scatterplot, bar chart, etc.) ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp)) + geom_point() # use a logscale if there's several orders of magnitude in one section # of the x-axis or y-axis ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp)) + geom_point() + scale_x_log10() # group the data by continent with color ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp, color = continent)) + geom_point() + scale_x_log10() # group the data by continent and size of population as aesthetics #3 and #4 # on top of aesthetics #1 (x) and #2 (y) ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp, color = continent, size = pop)) + geom_point() + scale_x_log10() ## Faceting # Faceting explores categorical variables into subplots. ggplot(gapminder_2007, aes(x = gdpPercap, y = lifeExp)) + geom_point() + scale_x_log10() + facet_wrap(~ continent) ## Summarize # finding powerful data points using summarize and its # functionality from the dplyr package # mean, min, max, sum and median are available functions # you can do as many of these functions as you wish gapminder %>% summarize(meanLifeExp = mean(lifeExp)) gapminder %>% filter(year == 2007) %>% summarize(meanLifeExp = mean(lifeExp)) gapminder %>% filter(year == 2007) %>% summarize(meanLifeExp = mean(lifeExp), totalPop = sum(pop)) ## group_By # group_by replaces filter and does the summarize by a "group" gapminder %>% group_by(year) %>% summarize(meanLifeExp = mean(lifeExp), totalPop = sum(pop)) # group_by is now done after filtering and does the summarize by a "group" gapminder %>% filter(year == 2007) %>% group_by(continent) %>% summarize(meanLifeExp = mean(lifeExp), totalPop = sum(pop)) # group_by is now done after filtering and does the summarize by a "group" gapminder %>% group_by(year, continent) %>% summarize(meanLifeExp = mean(lifeExp), totalPop = sum(pop)) ### Dynamic viz by_year <- gapminder %>% group_by(year) %>% summarize(totalPop = sum(pop), meanLifeExp = mean(lifeExp)) # plot these summaries now ggplot(by_year, aes(x = year, y = totalPop)) + geom_point() # add y = 0 to the graph ggplot(by_year, aes(x = year, y = totalPop)) + geom_point() + expand_limits(y = 0) # group by year, continent into new data frame by_year_continent <- gapminder %>% group_by(year, continent) %>% summarize(totalPop = sum(pop), meanLifeExp = mean(lifeExp)) ggplot(by_year_continent, aes(x = year, y = totalPop, color = continent)) + geom_point() + expand_limits(y = 0) ### Line Plots ggplot(by_year_continent, aes(x = year, y = meanLifeExp, color = continent)) + geom_line() + expand_limits(y = 0) ### Bar Plots # In a bar plot, the categorical goes on the x-axis # and the height of the bar is determined by the y-axis quantitative # variable by_continent <- gapminder %>% group_by(continent) %>% summarize(totalPop = sum(pop), meanLifeExp = mean(lifeExp)) ggplot(by_continent, aes(x = continent, y = meanLifeExp)) + geom_col() ### Histogram # Investigate one variable distribution at a time in "bins" # Uses only one axis in the aesthetic ggplot(gapminder_2007, aes(x = lifeExp)) + geom_histogram() # change the histogram bin width # you can also change the # of bins, using bins = <insert #> ggplot(gapminder_2007, aes(x = lifeExp)) + geom_histogram(binwidth = 5) ### Box Plots # Shows the importance of a variable distribution for comparison # box plots have 2 aesthetics where x = categorical and y = measurable # variable you want to compare ggplot(gapminder_2007, aes(x = continent, y = lifeExp)) + geom_boxplot() # add a label using labs and title ggplot(gapminder_2007, aes(x = lifeExp)) + geom_histogram(binwidth = 5) + labs( title = paste( "Life Expectancy Across Continents" ) ) # done anotehr way ggplot(gapminder_2007, aes(x = lifeExp)) + geom_histogram(binwidth = 5) + labs(title = "Life Expectancy Across Continents") # using ggtitle ggplot(gapminder_2007, aes(x = lifeExp)) + geom_histogram(binwidth = 5) + ggtitle("Life Expectancy Across Continents")
e5740c2a4b827baa6100263a18a3640728862c02
73139b16df978f4cd71e5d3a4e8820c62c9bdc67
/Analyze Data with R/Visualizing Carbon Dioxide Levels.R
71de12fb49e85e692a770f1a4db18c2ca773ee95
[]
no_license
oplt/Codecademy-Projects
9bbbc07f8c5957082baf1437ddfb4f5e3dd91cce
09a41bfb623d5d4859bfaaa8e7bc7d6461a9f7d4
refs/heads/main
2023-02-05T04:53:20.299992
2020-12-25T09:15:31
2020-12-25T09:15:31
null
0
0
null
null
null
null
UTF-8
R
false
false
1,166
r
Visualizing Carbon Dioxide Levels.R
--- title: "Visualizing Carbon Dioxide Levels" output: html_notebook --- # load libraries and data library(readr) library(dplyr) library(ggplot2) noaa_data <- read_csv("carbon_dioxide_levels.csv") head(noaa_data) options(scipen=10000) #removes scientific notation #Create NOAA Visualization here: noaa_viz <- ggplot(data = noaa_data, aes(x = Age_yrBP, y = CO2_ppmv)) + geom_line() + labs(title="Carbon Dioxide Levels From 8,000 to 136 Years BP", subtitle="From World Data Center for Paleoclimatology and NOAA Paleoclimatology Program", x="Years Before Today (0=1950)", y = "Carbon Dioxide Level (Parts Per Million)") + scale_x_reverse(lim=c(800000,0)) #Create IAC Visualization iac_data <- read_csv("yearly_co2.csv") head(iac_data) millennia_max <- max(noaa_data$CO2_ppmv) iac_viz <- ggplot(data=iac_data, aes(x=year, y=data_mean_global)) + geom_line() + labs(title = 'Carbon Dioxide Levels over Time', subtitle= 'From Institute for Atmospheric and Climate Science (IAC)', x="Year", y="Carbon Dioxide Level (Parts Per Million)") + geom_hline(aes(yintercept=millennia_max, linetype='Historical CO2 Peak before 1950'))
f659042ce4b54ed5826cf32c8b8e394ccba0a173
56146ed60db3ede66970748bd15757f901057092
/03_buffer_analysis.R
1c9ab373b6d0df6a6fa7c14d1d746e41bdfd8396
[]
no_license
annepaulus/FAWKES-III
d00ee66733e78b918aeb9fa8b378ed4e9cffb1fb
70095573faa183abbf84dd21c4269dcc0811321c
refs/heads/master
2021-05-24T08:07:30.214570
2020-01-31T09:51:38
2020-01-31T09:51:38
null
0
0
null
null
null
null
UTF-8
R
false
false
40,098
r
03_buffer_analysis.R
############################################################################ ### 4. overlay of SPAs with ACTs ############################################################################ ############################################################################### ### Buffer stuffer ############################################################################### # subset shapefile of 10 SPAs #subset<-Natura2000_shape_all[1:10,] subset<-Natura2000_shape_all #plot(subset, col="red", add=TRUE) # only subset for now #fun with outer buffers outerBuffer<-function(x, d){ buff<-buffer(x, width = d, dissolve = F) e<-erase(buff,x) return(e) } buffer10<-list() area<-list() perimeter<-list() # outer10km<-outerBuffer(x=subset,d=10000) myBuffer<-list() pb <- txtProgressBar(min = 0, max = length(subset), style = 3) # produces a progress bar for(i in 1:length(subset)){ myBuffer[[i]]<-outerBuffer(x=subset[i,],d=10000) setTxtProgressBar(pb, i) } close(pb) saveRDS(myBuffer, file = "myBuffer.rds") #plot(ACT) plot(myBuffer[[4]], col="red") pb <- txtProgressBar(min = 0, max = total, style = 3) # produces a progress bar for (i in 1:length(myBuffer)){ subset10 <- myBuffer[[i]] # takes each shapefile, if it is a SPA buffer10[[i]]<-extract(ACT,subset10,weights=TRUE,normalizeWeights=TRUE) area[[i]]<-gArea(subset10)/1e+06 perimeter[[i]]<-gLength(subset10)/1e+03 setTxtProgressBar(pb, i) } close(pb) #cleanup ACT_red<-list() area_red<-list() for (i in 1:length(buffer10)){ if(is.null(buffer10[[i]][[1]])==TRUE) { ACT_red[[i]]<-NA area_red[[i]]<-NA }else{ if(isTRUE(unique(is.na(as.data.frame(buffer10[[i]])$value)))==TRUE){ ACT_red[[i]]<-NA area_red[[i]]<-NA }else{ f.correct<-1/sum(as.data.frame(buffer10[[i]])[!is.na(as.data.frame(buffer10[[i]])[,1]),2]) reduced.na<-as.data.frame(buffer10[[i]])[!is.na(as.data.frame(buffer10[[i]])[,1]),] reduced.na[,2]<-reduced.na[,2]*f.correct ACT_red[[i]]<-reduced.na area_red[[i]]<-area[[i]]*sum(as.data.frame(buffer10[[i]])[!is.na(as.data.frame(buffer10[[i]])[,1]),2]) } } } # create a dataframe for presenting the distribution of SPA-area per ACT results <- as.data.frame(matrix(NA,length(ACT_red),20)) colnames(results) <- c("SITECODE","area",paste("ACT",c(1:17),sep=""),"count") for(k in 1:length(ACT_red)){ results[k,1]<-as.character(overlapSPA)[k] results[k,2]<-area_red[[k]] if(is.na(ACT_red[[k]])==TRUE){ results[k,3:19]<-NA } else{ test.sum<-cbind(aggregate(weight~value,sum,data=ACT_red[[k]]),table(ACT_red[[k]]$value)) results[k,test.sum$value+2]<-test.sum$weight results[k,20]<-sum(test.sum$Freq) } print(k) } results[is.na(results)]<-0 #results2 <- results[which(results$count>5),] saveRDS(results, file = "results_buffer.rds") ############################################################################### # DATA CLEANING # ############################################################################### # # get rid of NAs and compute class proportions # # ACT_red<-list() # area_red<-list() # # for (i in 1:total){ # if(is.null(ACT_ext[[i]][[1]])==TRUE) { # ACT_red[[i]]<-NA # area_red[[i]]<-NA # }else{ # if(isTRUE(unique(is.na(as.data.frame(ACT_ext[[i]])$value)))==TRUE){ # ACT_red[[i]]<-NA # area_red[[i]]<-NA # }else{ # f.correct<-1/sum(as.data.frame(ACT_ext[[i]])[!is.na(as.data.frame(ACT_ext[[i]])[,1]),2]) # reduced.na<-as.data.frame(ACT_ext[[i]])[!is.na(as.data.frame(ACT_ext[[i]])[,1]),] # reduced.na[,2]<-reduced.na[,2]*f.correct # ACT_red[[i]]<-reduced.na # area_red[[i]]<-area[[i]]*sum(as.data.frame(ACT_ext[[i]])[!is.na(as.data.frame(ACT_ext[[i]])[,1]),2]) # } # } # } # # # create a dataframe for presenting the distribution of SPA-area per ACT # results <- as.data.frame(matrix(NA,length(ACT_red),20)) # colnames(results) <- c("SITECODE","area",paste("ACT",c(1:17),sep=""),"count") # # for(k in 1:length(ACT_red)){ # results[k,1]<-as.character(overlapSPA)[k] # results[k,2]<-area_red[[k]] # # if(is.na(ACT_red[[k]])==TRUE){ # results[k,3:19]<-NA # } else{ # test.sum<-cbind(aggregate(weight~value,sum,data=ACT_red[[k]]),table(ACT_red[[k]]$value)) # results[k,test.sum$value+2]<-test.sum$weight # results[k,20]<-sum(test.sum$Freq) # # } # print(k) # } # # # # results[is.na(results)]<-0 # results2 <- results[which(results$count>5),] ############################################################################ ### relate conservation status with ACT via bird species ### ############################################################################ birds<-(unique(mydata$SPECIESNAME)) length(unique(mydata$SPECIESCODE)) # 530 bird species #for each species, link the ACTs (1 to 17) with the Conservation Status (for all sites tabFinal <- merge(mydata,results, by="SITECODE") tabFinal<-tabFinal[!duplicated(tabFinal), ] tabFinal[is.na(tabFinal)] <- 0 #remove rows completely 0 #test<-apply(tabFinal[,7:23], MARGIN=1,FUN=sum) #zero<-which(test==0) #tabFinal<-tabFinal[-zero,] #CStatus ordered factor CStatus<-droplevels(as.factor(tabFinal$CONSERVATION)) CStatus <- factor(CStatus,levels = c("A","B","C"),ordered = TRUE) tabFinal$CONSERVATION<-CStatus #tabFinal<-tabFinal[-which(tabFinal$SPECIESNAME=="--NULL--"),] tabFinal$SPECIESNAME<-droplevels(tabFinal$SPECIESNAME) # bird<-bird[1:511,1:3] # colnames(bird)<-c("SPECIESCODE","migration","preference") #bird <- bird[which(bird$migration=="mainly resident"),] tabFinal_buffer <- merge(tabFinal,bird, by="SPECIESCODE") #table(bird$SPECIESCODE) #which(table(bird$SPECIESCODE)>1) #save.image(file = "FAWKES.RData") saveRDS(tabFinal, file = "tabFinal_buffer.rds") ############################################################################ ### ############################################################################ dat_buffer <- data.frame(tabFinal_buffer[,c(1:6,24)], Conversion = rowSums(tabFinal_buffer[,18:22]), Intens = rowSums(tabFinal_buffer[,c(7:10,17)]), De_intens = rowSums(tabFinal_buffer[,11:16]), Stabil = c(tabFinal_buffer[,23]), Int_crop = rowSums(tabFinal_buffer[,7:10]), Ext_crop = rowSums(tabFinal_buffer[,11:13]), Ext_pasture = rowSums(tabFinal_buffer[,14:16]), Int_wood = c(tabFinal_buffer[,17]), Cropland_loss = rowSums(tabFinal_buffer[,20:21]), Forest_gain = c(tabFinal_buffer[,18]), Forest_loss = c(tabFinal_buffer[,19]), Urban = c(tabFinal_buffer[,22]) ) tab_buffer <- merge(dat_buffer,bird, by="SPECIESCODE") for (i in 1:length(dat_buffer$SPECIESCODE)){ dat_buffer$ACT_dom4[i]<-colnames(dat_buffer[i,8:11])[max.col(dat_buffer[i,8:11])] } for (i in 1:length(dat_buffer$SPECIESCODE)){ dat_buffer$ACT_dom9[i]<-colnames(dat_buffer[i,11:19])[max.col(dat_buffer[i,11:19])] } saveRDS(dat_buffer, file = "dat_buffer.rds") saveRDS(tab_buffer, file = "tab_buffer.rds") # # # bird2<-reshape(aggregate (dat$ACT_dom, list(dat$SPECIESCODE,dat$ACT_dom), FUN=length), v.names="x", timevar="Group.2", idvar="Group.1", direction="wide") # colnames(bird2)<-c("SPECIESCODE","Conversion", "De_intens" ,"Intens" ,"Stabil") # bird<-merge(bird,bird2, by="SPECIESCODE") # # colnames(bird)<-c("SPECIESCODE", "migration" ,"preference","Conversion", "De_intens" ,"Intens" ,"Stabil") # # for (i in 1:length(bird$SPECIESCODE)){ # # bird$ACT_dom_dom[i]<-colnames(bird[i,4:7])[max.col(bird[i,4:7])] # # } ######################################### #### Ordered logistic regression analysis for habitat Conservation status (ABC) against levers trajectories (4 summarized) #### This is using the extracted values from Levers in 10km buffers around all SPAs #### inspiration for code found at https://stats.idre.ucla.edu/r/dae/ordinal-logistic-regression/ #### Authors: MB, TV ... ######################################### #### re-calculate actual area, unit unknown ... dat_buffer$cc_area<-dat_buffer$area*dat_buffer$Conversion dat_buffer$int_area<-dat_buffer$area*dat_buffer$Intens dat_buffer$ext_area<-dat_buffer$area*dat_buffer$De_intens dat_buffer$stab_area<-dat_buffer$area*dat_buffer$Stabil #### create new dataframe, each species in one row bird_OR_results <- data.frame(unique(droplevels(dat_buffer$SPECIESCODE)), OR_int=seq(1:486), p_int=seq(1:486), LCI_OR_int=seq(1:486), UCI_OR_int=seq(1:486), OR_ext=seq(1:486), p_ext=seq(1:486), LCI_OR_ext=seq(1:486), UCI_OR_ext=seq(1:486), OR_cc=seq(1:486), p_cc=seq(1:486), LCI_OR_cc=seq(1:486), UCI_OR_cc=seq(1:486), OR_stab=seq(1:486), p_stab=seq(1:486), LCI_OR_stab=seq(1:486), UCI_OR_stab=seq(1:486), OR_Int_crop=seq(1:486), p_Int_crop=seq(1:486), LCI_OR_Int_crop=seq(1:486), UCI_OR_Int_crop=seq(1:486), OR_Ext_crop=seq(1:486), p_Ext_crop=seq(1:486), LCI_OR_Ext_crop=seq(1:486), UCI_OR_Ext_crop=seq(1:486), OR_Ext_pasture=seq(1:486), p_Ext_pasture=seq(1:486), LCI_OR_Ext_pasture=seq(1:486), UCI_OR_Ext_pasture=seq(1:486), OR_Int_wood=seq(1:486), p_Int_wood=seq(1:486), LCI_OR_Int_wood=seq(1:486), UCI_OR_Int_wood=seq(1:486), OR_Cropland_loss=seq(1:486), p_Cropland_loss=seq(1:486), LCI_OR_Cropland_loss=seq(1:486), UCI_OR_Cropland_loss=seq(1:486), OR_Forest_gain=seq(1:486), p_Forest_gain=seq(1:486), LCI_OR_Forest_gain=seq(1:486), UCI_OR_Forest_gain=seq(1:486), OR_Forest_loss=seq(1:486), p_Forest_loss=seq(1:486), LCI_OR_Forest_loss=seq(1:486), UCI_OR_Forest_loss=seq(1:486), OR_Urban=seq(1:486), p_Urban=seq(1:486), LCI_OR_Urban=seq(1:486), UCI_OR_Urban=seq(1:486), stringsAsFactors=FALSE) names(bird_OR_results)[1]<-"SPECIESCODE" #### loop running individual models for all bird species for (i in 1:length(bird_OR_results$SPECIESCODE)){ one_bird<-dat_buffer[dat_buffer$SPECIESCODE==paste(bird_OR_results$SPECIESCODE[i]),] one_bird<-(one_bird[(rowSums((one_bird[,11:19]))==0)<=0,]) # skip all species that occur in less than 5 SPAs if (nrow(one_bird) < 5){ bird_OR_results[i,2:9]<-NA } else { #### Intens models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Intens+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,2]<-res[1,1] bird_OR_results[i,4]<-res[1,2] bird_OR_results[i,5]<-res[1,3] bird_OR_results[i,3]<-ctable[1,4] #### De_intens models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~De_intens+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,6]<-res[1,1] bird_OR_results[i,8]<-res[1,2] bird_OR_results[i,9]<-res[1,3] bird_OR_results[i,7]<-ctable[1,4] #### Conversion models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Conversion+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,10]<-res[1,1] bird_OR_results[i,12]<-res[1,2] bird_OR_results[i,13]<-res[1,3] bird_OR_results[i,11]<-ctable[1,4] #### Stabil models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Stabil+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,14]<-res[1,1] bird_OR_results[i,16]<-res[1,2] bird_OR_results[i,17]<-res[1,3] bird_OR_results[i,15]<-ctable[1,4] #### Int_crop models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Int_crop+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,18]<-res[1,1] bird_OR_results[i,20]<-res[1,2] bird_OR_results[i,21]<-res[1,3] bird_OR_results[i,19]<-ctable[1,4] #### Ext_crop models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Ext_crop+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,22]<-res[1,1] bird_OR_results[i,24]<-res[1,2] bird_OR_results[i,25]<-res[1,3] bird_OR_results[i,23]<-ctable[1,4] #### Ext_pasture models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Ext_pasture+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,26]<-res[1,1] bird_OR_results[i,28]<-res[1,2] bird_OR_results[i,29]<-res[1,3] bird_OR_results[i,27]<-ctable[1,4] #### Int_wood models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Int_wood+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,30]<-res[1,1] bird_OR_results[i,32]<-res[1,2] bird_OR_results[i,33]<-res[1,3] bird_OR_results[i,31]<-ctable[1,4] #### Cropland_loss models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Cropland_loss+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,34]<-res[1,1] bird_OR_results[i,36]<-res[1,2] bird_OR_results[i,37]<-res[1,3] bird_OR_results[i,35]<-ctable[1,4] #### Forest_gain models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Forest_gain+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,38]<-res[1,1] bird_OR_results[i,40]<-res[1,2] bird_OR_results[i,41]<-res[1,3] bird_OR_results[i,39]<-ctable[1,4] #### Forest_loss models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Forest_loss+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,42]<-res[1,1] bird_OR_results[i,44]<-res[1,2] bird_OR_results[i,45]<-res[1,3] bird_OR_results[i,43]<-ctable[1,4] #### Urban models tryCatch({ m<-polr(formula=factor(one_bird$CONSERVATION,levels = c("C","B","A"),ordered = TRUE)~Urban+area, data=one_bird,Hess=TRUE) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) tryCatch({ ctable <- coef(summary(m)) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) tryCatch({ ci <- confint(m) }, error=function(e){cat("ERROR :",conditionMessage(e), "\n")}) #confint.default(m) #exp(coef(m)) res<-exp(cbind(OR = coef(m), ci)) bird_OR_results[i,46]<-res[1,1] bird_OR_results[i,48]<-res[1,2] bird_OR_results[i,49]<-res[1,3] bird_OR_results[i,47]<-ctable[1,4] } print(i) } saveRDS(bird_OR_results, file = "bird_OR_results_buffer.rds") bird_OR_results<-readRDS(file = "bird_OR_results_buffer.rds") #### results summary & plot int_neg<-summary((bird_OR_results$OR_int<1 & bird_OR_results$p_int<0.05)) int_pos<-summary((bird_OR_results$OR_int>1 & bird_OR_results$p_int<0.05)) ext_neg<-summary((bird_OR_results$OR_ext<1 & bird_OR_results$p_ext<0.05)) ext_pos<-summary((bird_OR_results$OR_ext>1 & bird_OR_results$p_ext<0.05)) cc_neg<-summary((bird_OR_results$OR_cc<1 & bird_OR_results$p_cc<0.05)) cc_pos<-summary((bird_OR_results$OR_cc>1 & bird_OR_results$p_cc<0.05)) stab_neg<-summary((bird_OR_results$OR_stab<1 & bird_OR_results$p_stab<0.05)) stab_pos<-summary((bird_OR_results$OR_stab>1 & bird_OR_results$p_stab<0.05)) Int_crop_neg<-summary((bird_OR_results$OR_Int_crop<1 & bird_OR_results$p_Int_crop<0.05)) Int_crop_pos<-summary((bird_OR_results$OR_Int_crop>1 & bird_OR_results$p_Int_crop<0.05)) Ext_crop_neg<-summary((bird_OR_results$OR_Ext_crop<1 & bird_OR_results$p_Ext_crop<0.05)) Ext_crop_pos<-summary((bird_OR_results$OR_Ext_crop>1 & bird_OR_results$p_Ext_crop<0.05)) Ext_pasture_neg<-summary((bird_OR_results$OR_Ext_pasture<1 & bird_OR_results$p_Ext_pasture<0.05)) Ext_pasture_pos<-summary((bird_OR_results$OR_Ext_pasture>1 & bird_OR_results$p_Ext_pasture<0.05)) Int_wood_neg<-summary((bird_OR_results$OR_Int_wood<1 & bird_OR_results$p_Int_wood<0.05)) Int_wood_pos<-summary((bird_OR_results$OR_Int_wood>1 & bird_OR_results$p_Int_wood<0.05)) Cropland_loss_neg<-summary((bird_OR_results$OR_Cropland_loss<1 & bird_OR_results$p_Cropland_loss<0.05)) Cropland_loss_pos<-summary((bird_OR_results$OR_Cropland_loss>1 & bird_OR_results$p_Cropland_loss<0.05)) Forest_gain_neg<-summary((bird_OR_results$OR_Forest_gain<1 & bird_OR_results$p_Forest_gain<0.05)) Forest_gain_pos<-summary((bird_OR_results$OR_Forest_gain>1 & bird_OR_results$p_Forest_gain<0.05)) Forest_loss_neg<-summary((bird_OR_results$OR_Forest_loss<1 & bird_OR_results$p_Forest_loss<0.05)) Forest_loss_pos<-summary((bird_OR_results$OR_Forest_loss>1 & bird_OR_results$p_Forest_loss<0.05)) Urban_neg<-summary((bird_OR_results$OR_Urban<1 & bird_OR_results$p_Urban<0.05)) Urban_pos<-summary((bird_OR_results$OR_Urban>1 & bird_OR_results$p_Urban<0.05)) res_summary<-rbind(int_neg,int_pos,ext_neg,ext_pos,cc_neg,cc_pos,stab_neg,stab_pos) res_summary<-rbind(Int_crop_neg,Int_crop_pos,Ext_crop_neg,Ext_crop_pos,Ext_pasture_neg,Ext_pasture_pos,Int_wood_neg,Int_wood_pos, Cropland_loss_neg,Cropland_loss_pos,Forest_gain_neg,Forest_gain_pos,Forest_loss_neg,Forest_loss_pos,Urban_neg,Urban_pos,stab_neg,stab_pos) res_summary<-res_summary[,-1] barplot(as.numeric(res_summary[,2]),names.arg=c(rownames(res_summary)), col=c("red","green"), las=2) bird_OR_results<-merge(bird_OR_results,bird, by = "SPECIESCODE") bird_pref<-unique(bird_OR_results$preference) for (i in 1:length(bird_pref)){ sub_bird<-bird_OR_results[bird_OR_results$preference==bird_pref[i],] ext_neg<-summary((sub_bird$OR_ext<1 & sub_bird$p_ext<0.05)) ext_pos<-summary((sub_bird$OR_ext>1 & sub_bird$p_ext<0.05)) cc_neg<-summary((sub_bird$OR_cc<1 & sub_bird$p_cc<0.05)) cc_pos<-summary((sub_bird$OR_cc>1 & sub_bird$p_cc<0.05)) stab_neg<-summary((sub_bird$OR_stab<1 & sub_bird$p_stab<0.05)) stab_pos<-summary((sub_bird$OR_stab>1 & sub_bird$p_stab<0.05)) Int_crop_neg<-summary((sub_bird$OR_Int_crop<1 & sub_bird$p_Int_crop<0.05)) Int_crop_pos<-summary((sub_bird$OR_Int_crop>1 & sub_bird$p_Int_crop<0.05)) Ext_crop_neg<-summary((sub_bird$OR_Ext_crop<1 & sub_bird$p_Ext_crop<0.05)) Ext_crop_pos<-summary((sub_bird$OR_Ext_crop>1 & sub_bird$p_Ext_crop<0.05)) Ext_pasture_neg<-summary((sub_bird$OR_Ext_pasture<1 & sub_bird$p_Ext_pasture<0.05)) Ext_pasture_pos<-summary((sub_bird$OR_Ext_pasture>1 & sub_bird$p_Ext_pasture<0.05)) Int_wood_neg<-summary((sub_bird$OR_Int_wood<1 & sub_bird$p_Int_wood<0.05)) Int_wood_pos<-summary((sub_bird$OR_Int_wood>1 & sub_bird$p_Int_wood<0.05)) Cropland_loss_neg<-summary((sub_bird$OR_Cropland_loss<1 & sub_bird$p_Cropland_loss<0.05)) Cropland_loss_pos<-summary((sub_bird$OR_Cropland_loss>1 & sub_bird$p_Cropland_loss<0.05)) Forest_gain_neg<-summary((sub_bird$OR_Forest_gain<1 & sub_bird$p_Forest_gain<0.05)) Forest_gain_pos<-summary((sub_bird$OR_Forest_gain>1 & sub_bird$p_Forest_gain<0.05)) Forest_loss_neg<-summary((sub_bird$OR_Forest_loss<1 & sub_bird$p_Forest_loss<0.05)) Forest_loss_pos<-summary((sub_bird$OR_Forest_loss>1 & sub_bird$p_Forest_loss<0.05)) Urban_neg<-summary((sub_bird$OR_Urban<1 & sub_bird$p_Urban<0.05)) Urban_pos<-summary((sub_bird$OR_Urban>1 & sub_bird$p_Urban<0.05)) res_summary<-rbind(int_neg,int_pos,ext_neg,ext_pos,cc_neg,cc_pos,stab_neg,stab_pos) res_summary<-rbind(Int_crop_neg,Int_crop_pos,Ext_crop_neg,Ext_crop_pos,Ext_pasture_neg,Ext_pasture_pos,Int_wood_neg,Int_wood_pos, Cropland_loss_neg,Cropland_loss_pos,Forest_gain_neg,Forest_gain_pos,Forest_loss_neg,Forest_loss_pos,Urban_neg,Urban_pos,stab_neg,stab_pos) res_summary<-res_summary[,-1] png(gsub("[[:punct:]]", " ", paste(bird_pref[i],"outside SPAs.png"))) par(mar=c(10,3,3,3)) barplot(as.numeric(res_summary[,2]),names.arg=c(rownames(res_summary)), col=c("red","green"), las=2, main=paste(bird_pref[i],"outside SPAs")) dev.off() } ##### Guy-inspired plot pcols<-c(15,19,23,27,31,35,39,43,47) plot_list = list() for (i in 1:length(pcols)){ sub<-bird_OR_results[bird_OR_results[pcols[i]]<0.05,] #sub<-sub[sub[pcols[i]+2]<10000,] #sub<-sub[sub[pcols[i]-1]>0.00001,] sub<-sub[order(sub[pcols[i]-1]),] sub$SPECIESCODE<-factor(sub$SPECIESCODE,levels=sub$SPECIESCODE) sub$SPECIESCODE<-droplevels(sub$SPECIESCODE) p = ggplot(sub[sub[pcols[i]-1]<10,], aes_string(x = colnames(sub[1]), y = colnames(sub[pcols[i]-1]))) + geom_point(size = 4) + geom_hline(yintercept=1)+ ylim(0,25)+ geom_errorbar(aes_string(ymax = colnames(sub[pcols[i]+1]), ymin = colnames(sub[pcols[i]+2]))) # plot_list[[i]] = p ggsave(p, file=paste0("plot_", colnames(sub[pcols[i]-1]),".png")) } ############################################################################ ### Derichlet-Regression ############################################################################ ### aggregation of data on either site or species level (both offer alternative ways to interpret the data) ### Aggregate on site level. We'll keep 1 entry per site giving us the proportion of Levers trajectories ("int","ext","cc","stab") within each site as well as the proportion of conservation classes of ALL birds within each site (ABC) + area of site and levers pixel count sites<-unique(tab$SITECODE) mat_sites<-as.data.frame(matrix(NA,length(sites), 9)) colnames(mat_sites)<-c("A","B","C","int","ext","cc","stab","area", "count") for(j in 1:length(sites)){ sub<-subset(tab,tab$SITECODE==sites[j]) prop.Cstat<-table(sub$CONSERVATION)/sum(table(sub$CONSERVATION)) mat_sites[j,1:3]<-prop.Cstat mat_sites[j,4:7]<-sub[1,8:11] mat_sites[j,8]<-sub[1,6] mat_sites[j,9]<-sub[1,7] # count } ### Aggregate on species level. We'll keep 1 entry per species giving us the proportion of Levers trajectories ("int","ext","cc","stab") the species encounters within all sites. This is done as a two-step aggregation: first for each site the species occurs in the majority trajectory is selected and second, these majority trajectories are summarized as proportions across all sites the species occurs in. In addition we have the proportion of conservation classes of that species within all sites (ABC). species<-unique(tab$SPECIESCODE) mat_bird<-as.data.frame(matrix(NA,length(species), 10)) colnames(mat_bird)<-c("A","B","C","int","ext","cc","stab", "SPECIESCODE","migration","preference") for(j in 1:length(species)){ sub<-subset(tab,tab$SPECIESCODE==species[j]) lu_agg<-factor() levels(lu_agg)<-c("Conversion","Stabil","De_intens","Intens") for(k in 1:length(sub$SPECIESCODE)){ lu_agg[k]<-names(which.max(sub[k,8:11])) } mat_bird[j,4:7]<-table(lu_agg)/sum(table(lu_agg)) prop.Cstat<-table(sub$CONSERVATION)/sum(table(sub$CONSERVATION)) mat_bird[j,1:3]<-prop.Cstat mat_bird[j,4:7]<-sub[1,8:11] mat_bird[j,8]<-as.character(sub[1,1]) mat_bird[j,9]<-as.character(sub[1,12]) mat_bird[j,10]<-as.character(sub[1,13]) } #remove all birds that have no levers data at all (~50) mat_bird<-mat_bird[rowSums((mat_bird[,4:7])==0)<=0,] ### Site-based Drichlet analysis ABC_status <- DR_data(mat_sites[, 1:3]) plot(ABC_status, cex = 0.5, a2d = list(colored = FALSE, c.grid = FALSE)) plot(rep(mat_sites$int, 3), as.numeric(ABC_status), pch = 21,cex=0.2, bg = rep(c("#E495A5", "#86B875", "#7DB0DD"), each = 39), xlab = "intensity", ylab = "Proportion",ylim = 0:1) first_model <- DirichReg(ABC_status ~ mat_sites$int+mat_sites$cc+mat_sites$ext) summary(first_model) first_model <- DirichReg(ABC_status ~ mat_sites$int) plot(rep(mat_sites$int, 3), as.numeric(ABC_status), pch = 21,cex=0.2, bg = rep(c("#E495A5", "#86B875", "#7DB0DD"), each = 39), xlab = "intensity", ylab = "Proportion",ylim = 0:1) Xnew <- data.frame(int = seq(min(mat_sites$int), max(mat_sites$int),length.out = 100)) for (i in 1:3) lines(cbind(Xnew, predict(first_model, Xnew)[, i]), col = c("#E495A5", "#86B875", "#7DB0DD")[i], lwd = 2) # pre<-predict(first_model) plot(pre[,1]~mat_sites$int) plot(pre[,2]~mat_sites$int) plot(pre[,3]~mat_sites$int) boxplot(mat_sites[,1:3]) ### Species-based Drichlet analysis for (i in 1:length(mat_bird$SPECIESCODE)){ mat_bird$ABC_dom[i]<-colnames(mat_bird[i,1:3])[max.col(mat_bird[i,1:3])] } for (i in 1:length(tab$SPECIESCODE)){ tab$ACT_dom[i]<-colnames(tab[i,8:11])[max.col(tab[i,8:11])] } dat$cc_area<-dat$area*dat$Conversion dat$int_area<-dat$area*dat$Intens dat$ext_area<-dat$area*dat$De_intens dat$stab_area<-dat$area*dat$Stabil one_bird<-dat[dat$SPECIESCODE=="A030",] #head(one_bird) one_bird$country<-factor(substr((one_bird$SITECODE),1,2)) #lapply(one_bird[, c("CONSERVATION","cc_area")], table) one_bird<-(one_bird[(rowSums((one_bird[,8:10]))==0)<=0,]) m<-polr(formula=CONSERVATION~ACT_dom+area, data=one_bird,Hess=TRUE) ctable <- coef(summary(m)) p <- pnorm(abs(ctable[, "t value"]), lower.tail = FALSE) * 2 ctable <- cbind(ctable, "p value" = p) ci <- confint(m) confint.default(m) exp(coef(m)) exp(cbind(OR = coef(m), ci)) summary(factor(one_bird$ACT_dom)) ggplot(one_bird, aes(x = CONSERVATION, y = int_area)) + geom_boxplot(size = .75) + geom_jitter(alpha = .5) + facet_grid(ACT_dom~country, margins = TRUE) + theme(axis.text.x = element_text(angle = 45, hjust = 1, vjust = 1)) sf <- function(y) { c('Y>=1' = qlogis(mean(y >= 1)), 'Y>=2' = qlogis(mean(y >= 2)), 'Y>=3' = qlogis(mean(y >= 3))) } s <- with(one_bird, summary(as.numeric(CONSERVATION) ~ ACT_dom + area, fun=sf)) glm(I(as.numeric(CONSERVATION) >= 2) ~ ACT_dom, family="binomial", data = one_bird) glm(I(as.numeric(CONSERVATION) >= 3) ~ ACT_dom, family="binomial", data = one_bird) s[, 4] <- s[, 4] - s[, 3] s[, 3] <- s[, 3] - s[, 3] plot(s, which=1:3, pch=1:3, xlab='logit', main=' ', xlim=range(s[,3:4])) farmland_birds <- c("A247","A110", "A255", "A257", "A025", "A133", "A243","A366", "A031","A348", "A377", "A376", "A379", "A382", "A096","A244","A245", "A251", "A338", "A339", "A341","A156","A242", "A383","A260","A278","A356", "A112","A357", "A275","A276", "A361","A210", "A352", "A351", "A309", "A128", "A232", "A142") trends_farmland <- read.delim("trends_farmland.txt", header=FALSE) trends_farmland <- cbind(farmland_birds, trends_farmland) colnames(trends_farmland) <- c("Speciescode", "Speciesname", "Baseyear", "Trend") farmland_trends <- as.vector(trends_farmland$Trend) forest_birds <- c("A086", "A256", "A263", "A104", "A365", "A335", "A334", "A373", "A207", "A454", "A238", "A240", "A236", "A542", "A321", "A332", "A342", "A344", "A328", "A327", "A326", "A325", "A274", "A313", "A315", "A314", "A234", "A372", "A318", "A317", "A362", "A332", "A165", "A287") trends_forest <- read.delim("trends_forest.txt", header=FALSE) trends_forest <- cbind(forest_birds, trends_forest) colnames(trends_forest) <- c("Speciescode", "Speciesname", "Baseyear", "Trend") forest_trends <- as.vector(trends_forest$Trend) trends_all <- read.delim("trends_all.txt", header=FALSE) colnames(trends_all) <- c("SPECIESCODE", "Speciesname", "Baseyear", "Trend") trends_farmland$type<-"farmland" trends_forest$type<-"forest" trends<-rbind(trends_farmland,trends_forest) colnames(trends)[1]<-"SPECIESCODE" mat_bird$SPECIESCODE<-factor(mat_bird$SPECIESCODE) mat_bird_trends<-merge(mat_bird,trends,by="SPECIESCODE") summary(lme(C~int+cc+Trend, random= ~ 1|SPECIESCODE,data=mat_bird_trends)) boxplot(A~Trend, data=mat_bird_trends) ############################################################################ ### 6. correlations between Conservation status and ACTs ############################################################################ ###Describ. Stat. boxplot(dat[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(tabFinal[,7:23],col=c("red","red","red","red","lightgreen","lightgreen", "lightgreen","lightgreen","lightgreen","lightgreen","red", "darkred","darkred","darkred","darkred","darkred", "green")) par(mfrow=c(3,1)) boxplot(subset(dat,dat$CONSERVATION=="A")[,c(8:11)],col="green") boxplot(subset(dat,dat$CONSERVATION=="B")[,c(8:11)],col="yellow") boxplot(subset(dat,dat$CONSERVATION=="C")[,c(8:11)],col="red") ##################### models for subgroup of birds ####################### prey <- subset(tab, tab$preference=="birds of prey") open_c <- subset(tab, tab$preference=="open land carnivores") open_h <- subset(tab, tab$preference=="open land herbivores" ) coastal <- subset(tab, tab$preference=="coastal species") lakes <- subset(tab, tab$preference=="wetland/lakes") wood_c <- subset(tab, tab$preference=="woodland carnivores" ) wood_h <- subset(tab, tab$preference=="woodland herbivores") par(mfrow=c(3,1)) boxplot(subset(open_h,open_c$CONSERVATION=="A")[,c(8:11)],col="green") boxplot(subset(open_h,open_c$CONSERVATION=="B")[,c(8:11)],col="yellow") boxplot(subset(open_h,open_c$CONSERVATION=="C")[,c(8:11)],col="red") par(mfrow=c(1,1)) boxplot(prey[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(open_c[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(open_h[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(wood_c[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(wood_h[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(lakes[,8:11],col=c("darkred","red","lightgreen","green")) boxplot(coastal[,8:11],col=c("darkred","red","lightgreen","green")) pom_prey <- vglm(factor(CONSERVATION,levels = c("A","B","C"),ordered = TRUE)~Conversion+Intens+De_intens+area,data=coastal, family = cumulative(parallel=FALSE)) pom.ll <- logLik(pom_prey) pom0 <- vglm(factor(CONSERVATION,levels = c("A","B","C"),ordered = TRUE)~1,data=coastal, family = cumulative(parallel=FALSE)) p0.ll <- logLik(pom0) # R^2 McFadden pom.mcfad <- as.vector(1 - (pom.ll/p0.ll)) pom.mcfad # prey: 0.00369, open_c: 0.0087, open_h: 0.004609, wood_c: 0.00709, wood_h: -5.73 ???, lakes: 0.01467, coastal: 0.00853 # counts > 5: prey: 0.0104, open_c: 0.0076, open_h: 0.0157, wood_c: 0.0131, wood_h: -2,09 ???, lakes: 0.00613, coastal: 0.0059 # counts > 5 & "mainly resident": prey: 0.01333, open_c: 0.0074, open_h: 0.02249, wood_c: 0.0188, wood_h: -2,2 ???, lakes: 0.00817, coastal: 0.00269 # "mainly resident": prey: 0.01046, open_c: 0.0076, open_h: 0.0157, wood_c: 0.01314, wood_h: 0.2263, lakes: 0.0061, coastal: 0.005969 a<-strsplit(as.character(dat$SPECIESNAME), " ") name<-c() for(i in 1:length(a)){ name[i] <-a[[i]][1] } pom0 <- vglm(factor(CONSERVATION,levels = c("A","B","C"),ordered = TRUE)~1,data=dat, family = cumulative(parallel=FALSE)) pom <- vglm(factor(CONSERVATION,levels = c("A","B","C"),ordered = TRUE)~Conversion+Intens+De_intens+area,data=dat, family = cumulative(parallel=FALSE)) ?vglm #plotvglm(pom) summary(pom) pom.ll <- logLik(pom) pom.ll p0.ll <- logLik(pom0) p0.ll # R^2 McFadden pom.mcfad <- as.vector(1 - (pom.ll/p0.ll)) pom.mcfad # R^2 Cox&Snell N <- length(dat[, 1]) # Anzahl der F?lle pom.cox <- as.vector(1 - exp((2/N) * (p0.ll - pom.ll))) pom.cox # R^2 Nagelkerke pom.nagel <- as.vector((1 - exp((2/N) * (p0.ll - pom.ll)))/(1 - exp(p0.ll)^(2/N))) pom.nagel #Functions for PseudoR2s RsqGLM <- function(obs = NULL, pred = NULL, model = NULL) { # version 1.2 (3 Jan 2015) model.provided <- ifelse(is.null(model), FALSE, TRUE) if (model.provided) { if (!("glm" %in% class(model))) stop ("'model' must be of class 'glm'.") if (!is.null(pred)) message("Argument 'pred' ignored in favour of 'model'.") if (!is.null(obs)) message("Argument 'obs' ignored in favour of 'model'.") obs <- model$y pred <- model$fitted.values } else { # if model not provided if (is.null(obs) | is.null(pred)) stop ("You must provide either 'obs' and 'pred', or a 'model' object of class 'glm'") if (length(obs) != length(pred)) stop ("'obs' and 'pred' must be of the same length (and in the same order).") if (!(obs %in% c(0, 1)) | pred < 0 | pred > 1) stop ("Sorry, 'obs' and 'pred' options currently only implemented for binomial GLMs (binary response variable with values 0 or 1) with logit link.") logit <- log(pred / (1 - pred)) model <- glm(obs ~ logit, family = "binomial") } null.mod <- glm(obs ~ 1, family = family(model)) loglike.M <- as.numeric(logLik(model)) loglike.0 <- as.numeric(logLik(null.mod)) N <- length(obs) # based on Nagelkerke 1991: CoxSnell <- 1 - exp(-(2 / N) * (loglike.M - loglike.0)) Nagelkerke <- CoxSnell / (1 - exp((2 * N ^ (-1)) * loglike.0)) # based on Allison 2014: McFadden <- 1 - (loglike.M / loglike.0) Tjur <- mean(pred[obs == 1]) - mean(pred[obs == 0]) sqPearson <- cor(obs, pred) ^ 2 return(list(CoxSnell = CoxSnell, Nagelkerke = Nagelkerke, McFadden = McFadden, Tjur = Tjur, sqPearson = sqPearson)) }
0050413a4eaac06037edad5bcb81196c32005068
2099a2b0f63f250e09f7cd7350ca45d212e2d364
/DUC-Dataset/Summary_p100_R/D092.AP900626-0010.html.R
7667b154b46ee1a28d75ca4552cb77fddf78db6d
[]
no_license
Angela7126/SLNSumEval
3548301645264f9656b67dc807aec93b636778ef
b9e7157a735555861d2baf6c182e807e732a9dd6
refs/heads/master
2023-04-20T06:41:01.728968
2021-05-12T03:40:11
2021-05-12T03:40:11
366,429,744
3
0
null
null
null
null
UTF-8
R
false
false
989
r
D092.AP900626-0010.html.R
<html> <head> <meta name="TextLength" content="SENT_NUM:6, WORD_NUM:113"> </head> <body bgcolor="white"> <a href="#0" id="0">After the June 4, 1989, military crackdown on the democracy movement, Fang and Li obtained refuge in the U.S. Embassy.</a> <a href="#1" id="1">Americans who rushed to donate money to help victims of the Armenian earthquake 18 months ago are responding more slowly to last week's killer earthquake in Iran.</a> <a href="#2" id="2">Iran is believede-scale structure in the universe.</a> <a href="#3" id="3">Colenso said, however, that contributions will ``surge upward'' as news spreads of the earthquake's destruction.</a> <a href="#4" id="4">AmeriCares, a private agency based in New Canaan, Conn., brought the first shipment of supplies from the United States into Iran on Sunday.</a> <a href="#5" id="5">AWARDS: National Award for Science and Technology, 1978; Chinese Academy of Sciences Award 1982; New York Academy of Sciences Award, 1988.</a> </body> </html>
f2e302bc6a9d4b7d0723ffe78ee0aec530bedc41
0cd778de37c5ce532f22bfe1363ea3820cf43582
/Taller Integración Numérica/Taller Integrales.R
b6d705a7270b447b98e6d9ed201e8ea2f8dfb890
[ "CC0-1.0" ]
permissive
GabrielGomez9898/Analisis-Numerico
873b413e481781152ff12f74f3464b8c6c9d6df0
b1bd6b5ccfd1d0102863371fbf09d4544a328105
refs/heads/master
2020-12-26T07:45:44.406075
2020-05-25T03:46:38
2020-05-25T03:46:38
237,437,384
1
0
null
null
null
null
UTF-8
R
false
false
6,194
r
Taller Integrales.R
#TALLER INTEGRALES rm(list=ls()) #PUNTO A ----------------------------------------------------------------------------- f<-function(x){ return(sin(2*x)) } trapecio<- function (a,b){ h = b-a trap = h*((f(b)+f(a))/2) return(trap) } trapecio (0,2) #PUNTO B ----------------------------------------------------------------------------- x<-c(0.1,0.2,0.3,0.4,0.5) y<-c(1.8,2.6,3.0,2.8,1.9) mat <- matrix(c(0,0,0,0,0,0), nrow=1,ncol = 5, byrow=TRUE) options(digits = 16) vec<-c(0) for(i in 1:length(x)){ for(j in 0:length(x)){ vec[j+1]=x[i]^j } mat<-rbind(mat,c(vec)) } mat <- mat[1:length(x)+1,] coef_pol <- (solve(mat,y)) #el solve solo sirve para matrices cuadradas cat("Resultado Polinomio: f(x)", coef_pol[5],"x^4 +",coef_pol[4],"X^3 +",coef_pol[3],"X^2 +",coef_pol[2],"X +",coef_pol[1]) f<-function(x){ return( coef_pol[5]*x^4 +coef_pol[4]*x^3 +coef_pol[3]*x^2 +coef_pol[2]*x +coef_pol[1]) } simpson <- function(vector_x){ #h=0.1 cont=1 secuencia_simpson=0 cont=1 const=4 for(i in 1:length(vector_x)){ if(cont==1){ secuencia_simpson=secuencia_simpson +f(vector_x[cont]) } if(cont > 1 && cont < length(vector_x)){ if(const==4){ secuencia_simpson=secuencia_simpson + 4*f(vector_x[cont]) const=2 }else{ secuencia_simpson=secuencia_simpson + 2*f(vector_x[cont]) const=4 } } if(cont==length(vector_x)){ secuencia_simpson=secuencia_simpson +f(vector_x[cont]) } cont=cont+1 } resultado=secuencia_simpson cat("Valor de la integral: ",resultado) return(resultado) } valorAprox=simpson(x) integrate(f,1.6,2) valorReal=27.44311111111179 errorRelativo= abs((valorReal-valorAprox)/valorAprox) cat("Error relativo: ",errorRelativo) #REGLA TRAPECIO trapecio<- function (a,b){ h = b-a trap = h*((f(b)+f(a))/2) return(trap) } trapecio (1.6,2) #PUNTO C ------------------------------------------------------------------------------- #Aplicar metodo de Simpson para solucionar la integral options(digits = 16) f<-function(x){ return( sqrt(1+cos(x)*cos(x))) } simpson <- function(a,b,n){ h = (b-a)/n xn<-c(0) cont=1 xn[cont]=0 cont=cont+1 for (i in 1:n) { xn[cont]= xn[cont-1]+h cont=cont+1 } secuencia_simpson=0 cont=1 const=4 for(i in 1:length(xn)){ if(cont==1){ secuencia_simpson=secuencia_simpson +f(xn[cont]) } if(cont > 1 && cont < length(xn)){ if(const==4){ secuencia_simpson=secuencia_simpson + 4*f(xn[cont]) const=2 }else{ secuencia_simpson=secuencia_simpson + 2*f(xn[cont]) const=4 } } if(cont==length(xn)){ secuencia_simpson=secuencia_simpson +f(xn[cont]) } cont=cont+1 } valor = (b-a)/(3*n) resultado=valor*secuencia_simpson cat("Valor de la integral: ",resultado) } simpson(0,2,4) #PUNTO D ------------------------------------------------------------------------------ #Utilizar cuadratura de Gauss para aproximar la integral options(digits = 16) f<- function(t){ return(t*exp(t)) } cuadratura_gauss<-function(){ m=2 # para encontrar un polinomio de grado <= 2 xk = sqrt(1/3) # Encontrar los xn despejando x en los polinomios de Legendre #Para encontrar los pesos Ck, como es de grado dos basta con 2 puntos, es decir 2 Xn y 2 Ck c1=(2*(1-(xk)^2))/(m^2*(xk)^2) c2=(2*(1-(-xk)^2))/(m^2*(-xk)^2) ck<-c(c1,c2) #Para generar el cambio de variable de t -> x de [1,2] a [-1,1] se tiene en cuenta #la formula (t-a)/(b-a) , resolviendo dicha fomrula obtenemos t = (x+1)/2 +1 # y reemplazamos t en la funcion original, de esta manera quedaria la nueva funcion: f_nueva<- function(x){ return(((x+1)/2)+1*exp((x+1)/2)+1) } cont=1 sumatoria_gauss=0 for (i in 1:length(ck)) { sumatoria_gauss=sumatoria_gauss+(ck[cont]*f_nueva(xk)) cont=cont+1 } cat("Solucion de integral por cuadrantes de Gauss: ",sumatoria_gauss) } cuadratura_gauss() #PUNTO E ------------------------------------------------------------------------------ options(digits = 16) cuadratura_gauss_partida<-function(){ m=2 # para encontrar un polinomio de grado <= 2 xk = sqrt(1/3) # Encontrar los xn despejando x en los polinomios de Legendre #Para encontrar los pesos Ck, como es de grado dos basta con 2 puntos, es decir 2 Xn y 2 Ck c1=(2*(1-(xk)^2))/(m^2*(xk)^2) c2=(2*(1-(-xk)^2))/(m^2*(-xk)^2) ck<-c(c1,c2) #Para generar el cambio de variable de t -> x de [1,1.5] a [-1,1] se tiene en cuenta #la formula (t-a)/(b-a) , resolviendo dicha fomrula obtenemos t = (0.5x+0.5)/2 +1 # y reemplazamos t en la funcion original, de esta manera quedaria la nueva funcion: f_nueva<- function(x){ return(((0.5*x+0.5)/2)+1*exp((0.5*x+0.5)/2)+1) } cont=1 sumatoria_gauss=0 for (i in 1:length(ck)) { sumatoria_gauss=sumatoria_gauss+(ck[cont]*f_nueva(xk)) cont=cont+1 } integral_1=sumatoria_gauss*(1/2) cat("integral 1: ",integral_1) #Para generar el cambio de variable de t -> x de [1.5,2] a [-1,1] se tiene en cuenta #la formula (t-a)/(b-a) , resolviendo dicha fomrula obtenemos t = (0.5x+0.5)/2 +1.5 # y reemplazamos t en la funcion original, de esta manera quedaria la nueva funcion: f_nueva<- function(x){ return(((0.5*x+0.5)/2)+1.5*exp((0.5*x+0.5)/2)+1.5) } cont=1 sumatoria_gauss=0 for (i in 1:length(ck)) { sumatoria_gauss=sumatoria_gauss+(ck[cont]*f_nueva(xk)) cont=cont+1 } integral_2=sumatoria_gauss*(1/2) cat("integral 2: ",integral_2) cat("\n") cat("Sumatoria de las dos particiones de integrales: ",integral_1+integral_2) } cuadratura_gauss_partida()
5c08a6cd5fbc3c91c3a2bc35e7c413b9c14bb5df
c0afe781ba59fd9bfe43ccf4819bf854fe0f683d
/주요 변수 양상.R
20c108e9edffb7bb911e82fbc26d87c0265c5105
[]
no_license
CHO-MINJUNG/BigContest2020
340d0605dbd2ded30aea620fd049a4be7278cb28
49faeeee3f3493bc0778d875091850ff4e47cda3
refs/heads/master
2022-12-11T04:12:17.914406
2020-09-15T17:24:35
2020-09-15T17:24:35
288,389,040
0
0
null
null
null
null
UHC
R
false
false
4,271
r
주요 변수 양상.R
library(fpp2) data = read.csv('C:/Users/chomjung/OneDrive - 명지대학교/빅콘테스트2020/1차_최종_데이터/1차_최종_최근_팀타자.csv', stringsAsFactors =TRUE, header = TRUE, encoding="UTF-8") NCdata= subset(data, 팀코드=='NC') SKdata= subset(data, 팀코드=='SK') KTdata= subset(data, 팀코드=='KT') OBdata= subset(data, 팀코드=='OB') WOdata= subset(data, 팀코드=='WO') HHdata= subset(data, 팀코드=='HH') SSdata= subset(data, 팀코드=='SS') LGdata= subset(data, 팀코드=='LG') HTdata= subset(data, 팀코드=='HT') LTdata= subset(data, 팀코드=='LT') # NC NCdata <- ts(NCdata$잔루,start = 2016,end = 2019,frequency = 20) NCdata = stl(NCdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(NCdata) ft = max(0,1-var(NCdata$time.series[,3])/(var(NCdata$time.series[,2]+NCdata$time.series[,3]))); ft fs = max(0,1-var(NCdata$time.series[,3])/(var(NCdata$time.series[,1]+NCdata$time.series[,3]))); fs # SK SKdata <- ts(SKdata$2루타,start = 2016,end = 2019,frequency = 20) SKdata = stl(SKdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(SKdata) ft = max(0,1-var(SKdata$time.series[,3])/(var(SKdata$time.series[,2]+SKdata$time.series[,3]))); ft fs = max(0,1-var(SKdata$time.series[,3])/(var(SKdata$time.series[,1]+SKdata$time.series[,3]))); fs # KT KTdata <- ts(KTdata$단타,start = 2016,end = 2019,frequency = 20) KTdata = stl(KTdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(KTdata) ft = max(0,1-var(KTdata$time.series[,3])/(var(KTdata$time.series[,2]+KTdata$time.series[,3]))); ft fs = max(0,1-var(KTdata$time.series[,3])/(var(KTdata$time.series[,1]+KTdata$time.series[,3]))); fs # OB OBdata <- ts(OBdata$단타,start = 2016,end = 2019,frequency = 20) OBdata = stl(OBdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(OBdata) ft = max(0,1-var(OBdata$time.series[,3])/(var(OBdata$time.series[,2]+OBdata$time.series[,3]))); ft fs = max(0,1-var(OBdata$time.series[,3])/(var(OBdata$time.series[,1]+OBdata$time.series[,3]))); fs # WO WOdata <- ts(WOdata$단타,start = 2016,end = 2019,frequency = 20) WOdata = stl(WOdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(WOdata) ft = max(0,1-var(WOdata$time.series[,3])/(var(WOdata$time.series[,2]+WOdata$time.series[,3]))); ft fs = max(0,1-var(WOdata$time.series[,3])/(var(WOdata$time.series[,1]+WOdata$time.series[,3]))); fs # HH HHdata <- ts(HHdata$단타,start = 2016,end = 2019,frequency = 20) HHdata = stl(HHdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(HHdata) ft = max(0,1-var(HHdata$time.series[,3])/(var(HHdata$time.series[,2]+HHdata$time.series[,3]))); ft fs = max(0,1-var(HHdata$time.series[,3])/(var(HHdata$time.series[,1]+HHdata$time.series[,3]))); fs # SS SSdata <- ts(SSdata$득점권타수,start = 2016,end = 2019,frequency = 20) SSdata = stl(SSdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(SSdata) ft = max(0,1-var(SSdata$time.series[,3])/(var(SSdata$time.series[,2]+SSdata$time.series[,3]))); ft fs = max(0,1-var(SSdata$time.series[,3])/(var(SSdata$time.series[,1]+SSdata$time.series[,3]))); fs # LG LGdata <- ts(LGdata$득점권타수,start = 2016,end = 2019,frequency = 20) LGdata = stl(LGdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(LGdata) ft = max(0,1-var(LGdata$time.series[,3])/(var(LGdata$time.series[,2]+LGdata$time.series[,3]))); ft fs = max(0,1-var(LGdata$time.series[,3])/(var(LGdata$time.series[,1]+LGdata$time.series[,3]))); fs # HT HTdata <- ts(HTdata$득점권타수,start = 2016,end = 2019,frequency = 20) HTdata = stl(HTdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(HTdata) ft = max(0,1-var(HTdata$time.series[,3])/(var(HTdata$time.series[,2]+HTdata$time.series[,3]))); ft fs = max(0,1-var(HTdata$time.series[,3])/(var(HTdata$time.series[,1]+HTdata$time.series[,3]))); fs # LT LTdata <- ts(LTdata$득점권타수,start = 2016,end = 2019,frequency = 20) LTdata = stl(LTdata, t.window=13, s.window="periodic", robust=TRUE) autoplot(LTdata) ft = max(0,1-var(LTdata$time.series[,3])/(var(LTdata$time.series[,2]+LTdata$time.series[,3]))); ft fs = max(0,1-var(LTdata$time.series[,3])/(var(LTdata$time.series[,1]+LTdata$time.series[,3]))); fs
cdfa1da78ef8bf525427ad3c0096193350c8f715
ffdea92d4315e4363dd4ae673a1a6adf82a761b5
/data/genthat_extracted_code/Rlab/examples/skew.Rd.R
a4d19e696d3005563c5a9ee173b736c4a54aad4c
[]
no_license
surayaaramli/typeRrh
d257ac8905c49123f4ccd4e377ee3dfc84d1636c
66e6996f31961bc8b9aafe1a6a6098327b66bf71
refs/heads/master
2023-05-05T04:05:31.617869
2019-04-25T22:10:06
2019-04-25T22:10:06
null
0
0
null
null
null
null
UTF-8
R
false
false
230
r
skew.Rd.R
library(Rlab) ### Name: skew ### Title: 3rd moment skewness ratio ### Aliases: skew ### Keywords: univar ### ** Examples set.seed(1) x <- rexp(100) # Get skewness coefficient estimate for exponential distribution skew(x)
894b0b5d4ddf9b0be388c069ec51b97a1ecf7f4e
b9e54258e540f0a0447045729bb4eecb0e490426
/Bölüm 17 - Veri Modelleme Aşamaları ve Veri Ön İşleme/17.13 - Cold Deck Imputation.R
79dd04abf145c29d2f591b8b326e40423a4697fd
[]
no_license
sudedanisman/RUdemy
b36b67b9e875206a5424f33cc784fd13506f8d8d
28a9814706873f5d2e5985e4ba795354144d52c4
refs/heads/master
2023-01-30T01:54:26.321218
2020-12-14T11:36:00
2020-12-14T11:36:00
null
0
0
null
null
null
null
UTF-8
R
false
false
564
r
17.13 - Cold Deck Imputation.R
## Cold Deck Imputatiion df <- student_placement_data_with_NA View(df) index <- which(is.na(df$percentage.in.Algorithms)) index length(index) # certifications # workshops index[1] names(df) df[c("certifications" , "workshops")][index[1] , ] d <- subset(df , select = c("percentage.in.Algorithms" , "certifications" , "workshops"), subset = ( certifications == "python" & workshops == "data science" ) ) ort <- mean(d$percentage.in.Algorithms , na.rm = T) ort df$percentage.in.Algorithms[index[1]] <- ort df$percentage.in.Algorithms[31]
c95286baafd08433dbf140fc779b22e57d450bfa
e4c44366d606a27749d0d230e316ef46cf41d6af
/ImmoDaten/Immo_SAR_benchmark.R
eee9e05d53fb9904166cf0fdb9836164d5c88460
[]
no_license
JohannesJacob/APA-SatelliteImages
fe53a34b0301a05a9618f8f206743bce10104242
8487188ae1d3e2e8cb4da5ee5d2df4a96bc14853
refs/heads/master
2020-04-05T12:09:02.217440
2017-08-31T13:40:12
2017-08-31T13:40:12
95,221,425
0
0
null
null
null
null
UTF-8
R
false
false
1,738
r
Immo_SAR_benchmark.R
# Spatial Auto-Regression on Immo-Data library(HSAR) df_street <- read.csv("Immo_streetCoordinates.csv") df_street_noNA <- df_street[complete.cases(df_street),] # Convert data frame to a spatial object library(sp) spdf_street <- SpatialPointsDataFrame(coords = df_street_noNA[, c("lng", "lat")], proj4string = CRS("+proj=longlat +datum=WGS84 +ellps=WGS84 +towgs84=0,0,0"), data = df_street_noNA) coords <- coordinates(spdf_street) # extract the land parcel level spatial weights matrix library(spdep) nb.1.5 <- dnearneigh(coords, d1 = 0, d2 = 1.5 * max_1nn, row.names = IDs) knn.10 <- knearneigh(coords, k = 10) knn.10 <- knn2nb(knn.10) # to a weights matrix dist.1.5 <- nbdists(nb.1.5,spdf_street) # or knn.10 #dist.1.5 <- lapply(dist.25,function(x) exp(-0.5 * (x / 2500)^2)) mat.1.5 <- nb2mat(nb.1.5,glist=dist.1.5,style="W") W <- as(mat.1.5,"dgCMatrix") ## run the sar() function res.formula <- price ~. model <- lm(formula=res.formula,data=df_street_noNA) betas= coef(lm(formula=res.formula, data=df_street_noNA)) # minus NA coefficients betas= coef(lm(formula=res.formula, data=df_street_noNA[, -c(1:5,81, 92,118:121,128,130,134,137:140,143,146,149,152,155:158, 126, 162,165,168,171,174,177:180,186,189,192:199)])) pars=list(rho = 0.5, sigma2e = 2.0, betas = betas) res <- sar(res.formula,data=df_street_noNA[, -c(1:5,81,92,118:121,128,130,134,137:140,143,146,149,152,155:158, 126, 162,165,168,171,174,177:180,186,189,192:199)], W=W, burnin=5000, Nsim=10000, thinning=0, parameters.start=pars)#changed thinning to zero summary(res)
aad332a282e3fa84545887bf82eed7bcf4859241
3c4298be0c3b00e13690f11616097207e3215612
/man/reverse.Rd
97deecef5baeff681bbcfe9b90b9dfd8dd1e2dbe
[]
no_license
cran/quest
4ab4799f3d5a4f1fa980927ad7d7d7022e2aa2b6
0724147528574656cba132d423346d59514e5780
refs/heads/master
2023-08-03T09:11:28.856628
2021-09-10T10:20:02
2021-09-10T10:20:02
405,129,468
0
0
null
null
null
null
UTF-8
R
false
true
1,082
rd
reverse.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/quest_functions.R \name{reverse} \alias{reverse} \title{Reverse Code a Numeric Vector} \usage{ reverse(x, mini, maxi) } \arguments{ \item{x}{numeric vector.} \item{mini}{numeric vector of length 1 specifying the minimum numeric value.} \item{maxi}{numeric vector of length 1 specifying the maximum numeric value.} } \value{ numeric vector that correlates exactly -1 with \code{x}. } \description{ \code{reverse} reverse codes a numeric vector based on minimum and maximum values. For example, say numerical values of response options can range from 1 to 4. The function will change 1 to 4, 2 to 3, 3 to 2, and 4 to 1. If there are an odd number of response options, the middle in the sequence will be unchanged. } \examples{ x <- psych::bfi[[1]] head(x, n = 15) y <- reverse(x = psych::bfi[[1]], min = 1, max = 6) head(y, n = 15) cor(x, y, use = "complete.obs") } \seealso{ \code{\link{reverses}} \code{\link[psych]{reverse.code}} \code{\link[car]{recode}} }
23e4412d116d4c31ad20df8b9f790485707517d0
2ab96d6236c3e25ceb0a128e53bf025c413f04e8
/Week7/dataSketch.R
fcfa8dfe53b706a0554cc53bcca0c74c32318c75
[]
no_license
miamiww/ComputationalMedia
0eb66b206fb353ecc7497a8bb9ef7d0e41362a1b
f925983eb1a075cdc39444f17706b23c9a6eea10
refs/heads/master
2021-01-23T07:09:13.441410
2017-12-11T19:32:32
2017-12-11T19:32:32
102,499,677
0
0
null
null
null
null
UTF-8
R
false
false
63
r
dataSketch.R
airData <- read.csv("PRSA-adapted-aparrish.csv") str(airData)
bb9a7a903ab23aa43386d04ea6eef133a606fba4
b39f4a7acf1766383efbeafb4c10ef00cfdaafbe
/man/load_bps.Rd
4b7a7672991be58b42ae1198c0108feebce222de
[]
no_license
hmorzaria/atlantisom
6e7e465a20f3673d3dbc843f164fe211faddc6fb
b4b8bd8694c2537e70c748e62ce5c457113b796f
refs/heads/master
2020-04-02T02:24:05.672801
2015-12-12T00:06:03
2015-12-12T00:06:23
null
0
0
null
null
null
null
UTF-8
R
false
true
1,639
rd
load_bps.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/load_bps.R \name{load_bps} \alias{load_bps} \title{Extracts the names of the epibenthic biomasspools from the initial conditions file.} \usage{ load_bps(dir = getwd(), fgs, file_init) } \arguments{ \item{dir}{The directory of the atlantis model output, where the default is \code{getwd()}.} \item{fgs}{A data frame created by \code{\link{load_fgs}} that reads in the csv file containing functional group names, usually \code{"functionalGroups.csv"}.} \item{file_init}{A character value giving the file name of the intial conditions file. The file should be located in your current working directory or the folder you specify in \code{dir}. The argument can also be specified as the full path name, just as long as argument \code{dir} is specified as \code{NULL}. Usually the file is named \code{"init[...].nc".}, but it must end in \code{.nc}.} } \value{ A \code{vector} of biomass pools. } \description{ Use \code{fgs} \code{data.frame} as read in by \code{\link{load_fgs}} to get the biomass pool information. } \examples{ testscenario <- "INIT_VMPA_Jan2015" d <- system.file("extdata", testscenario, package = "atlantisom") file <- "functionalGroups.csv" fgs <- load_fgs(d, file) bps <- load_bps(dir = d, fgs = fgs, file_init = "INIT_VMPA_Jan2015.nc") } \author{ Alexander Keth } \seealso{ \code{\link{load_fgs}} Other load functions: \code{\link{load_biolprm}}, \code{\link{load_boxarea}}, \code{\link{load_box}}, \code{\link{load_diet_comp}}, \code{\link{load_fgs}}, \code{\link{load_meta}}, \code{\link{load_nc}}, \code{\link{load_runprm}} }
c484a9096223b0d739fbf5ce172808b2770d183d
d5ecf34f545c6dfffb920cdaad3b8c89bf23af74
/shiny/personnasIdentificateur/loadData.R
58f65e7128a73fa15e7d1ab16599692273769b07
[ "LPPL-1.3c" ]
permissive
davebulaval/Actulab_COOP
8b69bc082e28909354eef4c9aea75c5d605edc4e
4a6c652ac0d378ff4163aa10a1da5fa853839c6c
refs/heads/master
2021-08-30T08:09:21.487975
2017-12-16T23:17:17
2017-12-16T23:17:17
109,923,867
0
0
null
null
null
null
UTF-8
R
false
false
717
r
loadData.R
# Shapefile FSA.shape <- rgdal::readOGR(dsn = "shapefileFSA", layer = "grta000a11a_f") FSA.shapeQC <- subset(FSA.shape, substr(FSA.shape$RTACIDU, 1, 1) %in% c("G", "H", "J")) # Import dataset library(readr) dataSetPredict <- read_csv("DatasetModif/dataSetPredict.csv", col_types = cols(X1 = col_skip())) probColocation <- read_csv("DatasetModif/probColocation.csv", col_types = cols(X1 = col_skip())) dataEducation <- read_csv("DatasetModif/dataEducation.csv", col_types = cols(X1 = col_skip())) dataEmploi <- read_csv("DatasetModif/dataEmploi.csv", col_types = cols(X1 = col_skip()))
b680603f66715eb99e54a5f32534680f31618384
582436aa42fd1930c663c0b63ca440f1b96111ab
/3_Experiments_Stabilising_AHHA_SPACE/2_Iterative_Stabilisation/Scripts/Iterative_Stabilisation.R
5d7597a52f9139ff1a5f5b1e7cc052ed3e14f4c3
[]
no_license
JamesCollerton/MSc_Thesis
1543f2c5e498b536d90e7e5eb787f491cb317867
9ff52287102005f86dabc1230c1768564d6ac4d0
refs/heads/master
2020-05-25T10:57:45.341264
2017-03-26T10:06:23
2017-03-26T10:06:23
42,585,457
0
0
null
null
null
null
UTF-8
R
false
false
10,810
r
Iterative_Stabilisation.R
# - James Collerton # - Student Number 46114 # - Source Code for MSc Thesis # Stabilising the SPACE model by iteration. This script requires some manual # work as it is quicker to manually alter this script and use the results with # the Eureqa desktop application than to try and automate the whole pipeline # using the Eureqa API. Also, the Eureqa API is fairly limited in what it can # be applied to as it is limited in the number of variables/ amount of data # it can take in. ############################################################################### # Global Variables and settings. # Setting the current directory to where we have stored our data. setwd("~/Documents/Programming/Git_Folders/MSc-Thesis-Code/Submission/3_Experiments_Stabilising_AHHA_SPACE/2_Iterative_Stabilisation/Data") # Later this will be filled with the new, stable components. new_data = list() # Read these in either for comparison or to form part of the model. # We need unstable_df and data as the same thing, as one will be overwritten # throughout the program, whilst the other will not. unstable_df = read.csv("unstable_data.txt") comparison_data = read.csv("Training_Data.txt") data = read.csv("unstable_data.txt") # Initial conditions. N = 200 Q = data$Q[1] initial_scouts = data$S[1] total_passive = N - initial_scouts # Indicators for the data vectors. T_ind = 1 S_ind = 2 P0_ind = 3 P1_ind = 4 P2_ind = 5 A1_ind = 6 A2_ind = 7 R1_ind = 8 R2_ind = 9 ############################################################################### # Initialising # Initial conditions. These have been hardcoded for the single trial, but # could easily be changed to data$S[1], data$A1[1] etc. to allow flexibility. S = initial_scouts A_one = 0 A_two = 0 R_one = 0 R_two = 0 P_zero = total_passive P_one = 0 P_two = 0 # Initialising empty vectors to store the results in. S_vec = c() A_one_vec = c() A_two_vec =c() R_one_vec = c() R_two_vec = c() P_zero_vec = c() P_one_vec = c() P_two_vec = c() ############################################################################### # Differential functions dS <- function(S, t, Q, N){ result = 1.06669377213794e-9*t^3 - 0.0130383120110502 - 0.000168544171939409*t - 6.65198441305169e-13*t^4 - 2.20232908623747e-7*t^2 return(result) } dA_one <- function(S, P1, P2, R2, t, Q, N){ result = 0.00231744609341487*S + 0.942990934721937*R2* ( (R2 > 0) * 0.00157400238903202 + (R2 <= 0) * cos(0.855282848706793*S) ) - 0.0434850619886643 - 2.33968166891488e-5*S^2 - 0.00563945347163448*sin(5.19729925857688 + 0.364203873387924*S) return(result) } dA_two <- function(S, P1, P2, A1, t, Q, N){ result = 0.000282667076235126*A1 + 0.0950374171430637/S + 0.000297132846101472*A1*P2 + 0.030200635796381* (10.6088331903667 < S) - 0.0249260226030619 - 0.000370254067916185*P2 - 0.0042652733456114*P2* (10.6088331903667 < S) - 7.45572148229618e-6*S*P2*A1^2 return(result) } dR_one <- function(S, A1, A2, P1, P2, t, Q, N){ result = 0.0266282174935782 + 0.009329126372239*S + 0.0231123108033275*(P1 > 0) - 6.99557347486849e-5*t - 0.00535770880365551*A1 - 0.000343308507165349*S*A1 - 0.000231772856346287*S^2 return(result) } dR_two <- function(S, A1, A2, P1, P2, R1, t, Q, N){ result = 0.00410346545735221*S - 0.0110700090049996 - 9.0219126132179e-5* ( (S <= 6.52877096324257) * (-75.5959114648096) + (S > 6.52877096324257) * S )* ( (0.425348476983269 < P1) * 1.92739381263045 + (0.425348476983269 >= P1) * (S - 9.6421558706767*P1) ) return(result) } dP_one <- function(S, P2, t, Q, N){ result = 0.00011585819289194*P2* (72.3159612664517 <= P2)*cos(-0.0123745627255911*t) return(result) } dP_two <- function(S, t, Q, N){ result = 0.00235796547198259*t*exp(-0.255561049906936*S) - cos(S)*exp(-0.377989973882601*S) return(result) } ############################################################################### # Updating functions. # These functions are used to update the data vector to be written back in. # The idea is that each time we run the model, we stabilise a different component # (ex the number of scouts) and then use that stable part in declaring the # next function in Eureqa. update_S <- function(S_vec, data){ data$S <- S_vec return(data) } update_P2 <- function(P2_vec, data){ data$P2 <- P2_vec return(data) } update_P1 <- function(P1_vec, data){ data$P1 <- P1_vec return(data) } update_A1 <- function(A1_vec, data){ data$A1 <- A1_vec return(data) } update_A2 <- function(A2_vec, data){ data$A2 <- A2_vec return(data) } update_R1 <- function(R1_vec, data){ data$R1 <- R1_vec return(data) } update_R2 <- function(R2_vec, data){ data$R2 <- R2_vec return(data) } # This updates the training data file. For example, lets say that at the start # we have the variable S, and we define a function dS/dt = g(t, Q, N), then # we know that each time we run the simulation the S values won't change. We then # update the training data with these new stable S values, and build the next term # on top of that. So dA1/dt = h(S, t, Q, N). write_new_file <- function(new_data){ setwd("~/Documents/Programming/Git_Folders/MSc-Thesis-Code/Submission/3_Experiments_Stabilising_AHHA_SPACE/2_Iterative_Stabilisation/Data") S = c() P0 = c() P1 = c() P2 = c() A1 = c() A2 = c() R1 = c() R2 = c() N = c() Q = c() t = c() for(single_file in new_data){ S = c(S, single_file$S) P0 = c(P0, single_file$P0) P1 = c(P1, single_file$P1) P2 = c(P2, single_file$P2) A1 = c(A1, single_file$A1) A2 = c(A2, single_file$A2) R1 = c(R1, single_file$R1) R2 = c(R2, single_file$R2) N = c(N, single_file$N) Q = c(Q, single_file$Q) t = c(t, single_file$t) } write_df <- data.frame(S, A1, A2, P0, P1, P2, R1, R2, Q, t, N) write.csv(write_df, "new_data.txt", row.names = FALSE) } ############################################################################### # Running model with the underlying data included t = 1 while(t <= nrow(data)){ S_vec = c(S_vec, S) A_one_vec = c(A_one_vec, A_one) A_two_vec = c(A_two_vec, A_two) R_one_vec = c(R_one_vec, R_one) R_two_vec = c(R_two_vec, R_two) P_one_vec = c(P_one_vec, P_one) P_two_vec = c(P_two_vec, P_two) S = S + dS(S, t, Q, N) A_one = A_one + dA_one(S, P_one, P_two, R_two, t, Q, N) A_two = A_two + dA_two(S, P_one, P_two, A_one, t, Q, N) R_one = R_one + dR_one(S, A_one, A_two, P_one, P_two, t, Q, N) R_two = R_two + dR_two(S, A_one, A_two, P_one, P_two, R_one, t, Q, N) P_one = P_one + dP_one(S, P_two, t, Q, N) P_two = P_two + dP_two(S, t, Q, N) t = t + 1 } # Plotting the results plot_result <- function(vec, compare_data, unstable_data, y_label, leg_x, leg_y){ par(mfrow = c(1,2)) plot(compare_data, type = "l", col = "firebrick1", xlab = "Time", ylab = y_label, lwd = 2) lines(vec, col = "dodgerblue", lwd = 2) lines(unstable_data, col = "lightgreen", lwd = 2) legend(leg_x, leg_y, c("Underlying", "Stabilised", "Unstabilised"), lty=c(1, 1, 1), lwd=c(2, 2, 2), col=c("dodgerblue", "firebrick1", "lightgreen"), cex = 0.8) plot(compare_data - vec, type = "l", col = "red", lwd = 2.5, xlab = "Time", ylab = "Difference in Functions") } # Calculating the R squared value. R_squared <- function(estimate, real_values){ obs_mean = mean(estimate) SS_res = sum((estimate - real_values)^2) SS_tot = sum((real_values - obs_mean)^2) return(1 - SS_res/SS_tot) } # Mean squared error MSE <- function(estimate, real_values){ residuals = real_values - estimate return(sum(residuals^2) / length(residuals)) } # Mean Absolute Error MAE <- function(estimate, real_values){ abs_residuals = abs(real_values - estimate) return(sum(abs_residuals) / length(abs_residuals)) } # Creating vector of the R squared values. create_Rsq_vec <- function(){ c(S_rsq = R_squared(S_vec, comparison_data$S), A1_rsq = R_squared(A_one_vec, comparison_data$A1), A2_rsq = R_squared(A_two_vec, comparison_data$A2), R1_rsq = R_squared(R_one_vec, comparison_data$R1), R2_rsq = R_squared(R_two_vec, comparison_data$R2), P1_rsq = R_squared(P_one_vec, comparison_data$P1), P2_rsq = R_squared(P_two_vec, comparison_data$P2)) } # Creating vector of the MAE values. create_MAE_vec <- function(){ c(S_MAE = MAE(S_vec, comparison_data$S), A1_MAE = MAE(A_one_vec, comparison_data$A1), A2_MAE = MAE(A_two_vec, comparison_data$A2), R1_MAE = MAE(R_one_vec, comparison_data$R1), R2_MAE = MAE(R_two_vec, comparison_data$R2), P1_MAE = MAE(P_one_vec, comparison_data$P1), P2_MAE = MAE(P_two_vec, comparison_data$P2)) } # Creating vector of the MSE values. create_MSE_vec <- function(){ c(S_MSE = MSE(S_vec, comparison_data$S), A1_MSE = MSE(A_one_vec, comparison_data$A1), A2_MSE = MSE(A_two_vec, comparison_data$A2), R1_MSE = MSE(R_one_vec, comparison_data$R1), R2_MSE = MSE(R_two_vec, comparison_data$R2), P1_MSE = MSE(P_one_vec, comparison_data$P1), P2_MSE = MSE(P_two_vec, comparison_data$P2)) } ############################################################################### # Running the plotting of the results. plot_result(S_vec, comparison_data$S, unstable_df$S, "Number of Scouts", 0, 14) plot_result(A_one_vec, comparison_data$A1, unstable_df$A1, "Number of Assessors (Site One)", 0, 5.6) plot_result(A_two_vec, comparison_data$A2, unstable_df$A2, "Number of Assessors (Site Two)", 0, 4.6) plot_result(R_one_vec, comparison_data$R1, unstable_df$R1, "Number of Recruiters (Site One)", 0, 10.4) plot_result(R_two_vec, comparison_data$R2, unstable_df$R2, "Number of Recruiters (Site Two)", 0, 27.5) plot_result(P_one_vec, comparison_data$P1, unstable_df$P1, "Number of Passive (Site One)", 0, 1.5) plot_result(P_two_vec, comparison_data$P2, unstable_df$P2, "Number of Passive (Site Two)", 0, 133) ############################################################################### # Updating the values and preparing the new data vector to be written. data = update_S(S_vec, data) data = update_P2(P_two_vec, data) data = update_P1(P_one_vec, data) data = update_A1(A_one_vec, data) data = update_A2(A_two_vec, data) data = update_R1(R_one_vec, data) data = update_R2(R_two_vec, data) new_data[[length(new_data) + 1]] <- data ############################################################################### # Calculating the R squared values, MAE and MSE. create_Rsq_vec() create_MAE_vec() create_MSE_vec() ############################################################################### ############################################################################### # Write new files # write_new_file(new_data) ###############################################################################
6329639c8d75827b81a5ec9074c0c979b0b76151
6fec9d9c8a3718bd3e45a1db4d854732ef875a07
/LeadWorkerStudy/lead_data_strunc.R
fc13ca0c6d3530dd8ba843f52b8281e6111191d2
[]
no_license
bikram12345k/senstP
35884caf87724e129f5b3e9138ba88f38f66ef16
2b30a3d1a9c9ec5aadb4402f6ef8fe0bba71bbcd
refs/heads/master
2021-07-13T16:31:31.738108
2021-03-11T18:50:03
2021-03-11T18:50:03
239,898,297
0
0
null
null
null
null
UTF-8
R
false
false
3,902
r
lead_data_strunc.R
library(senstrat) library(sensitivitymw) library(sensitivitymv) library(struncatedP) library(partitions) library(foreach) library(doSNOW) lead <- read.csv('lead_absorption1.csv') gamSeq = seq(1, 5, .05) res_original = matrix(NA, length(gamSeq), 5) for(gamma in gamSeq){ pvals <- c() ## Exposed child vs control child pvals <- c(pvals, senmw(cbind(lead$lead[1:33], lead$lead[(33+1):(33+33)]), gamma=gamma, m1=7, m2=8, m=8)$pval) ## High and Medium vs Low exposure who <- lead$exposed_gp==1 st = rep(1, sum(who)) z = lead[who, 'job_exposure'] %in% c("h","m") sc = rscores(lead$lead[who], z, st, method="U788") pvals <- c(pvals, sen2sample(sc, z, alternative="greater", gamma=gamma )$pval) ## High vs Medium exposure who <- lead$exposed_gp==1 & lead$job_exposure != "l" z = lead[who, 'job_exposure'] == "h" st = rep(1, sum(who)) sc = rscores(lead$lead[who], z, st, method="U788") pvals <- c(pvals, sen2sample(sc, z, alternative="greater", gamma=gamma )$pval) ## Good vs Moderate or poor hygine who <- lead$exposed_gp==1 & lead$job_exposure == "h" st = rep(1, sum(who)) z = lead[who, 'hygiene'] != "g" sc = rscores(lead$lead[who], z, st, method="U788") pvals <- c(pvals, sen2sample(sc, z, alternative="greater", gamma=gamma )$pval) ## Moderate vs poor hygine who <- lead$exposed_gp==1 & lead$job_exposure == "h" & lead$hygiene != 'g' st = rep(1, sum(who)) z = lead[who, 'hygiene'] == "p" sc = rscores(lead$lead[who], z, st, method="U788") pvals <- c(pvals, sen2sample(sc, z, alternative="greater", gamma=gamma )$pval) #pvals res_original[which(gamma==gamSeq),] = pvals n = length(pvals) } #round(res, 6) rownames(res_original) = gamSeq #round(res_original, 6) foo <- function(f, trunc=.15) function(x) ifelse(x<.5,f((x-trunc)/(.5-trunc)),0) ## Table 4 tab = round(t(apply(res_original[,c(1,2,3,4,5)], 1, function(p) c( (5)*min(tail(sort(p), 5)/(1:(5))), sort(p)[1]*(5), pchisq(-2*log(prod(tail(sort(p), 5))), 2*(5), lower.tail=FALSE), truncatedP(p, trunc=.2), getstruncatedPval1(p, xi=foo(xiVp, trunc=.2), trunc=.2, N=20)) )), 5) rownames(tab) = gamSeq colnames(tab) = c('Simes', 'Holm', 'Fisher', 'truncatedP', 'struncatedP') tab ########################################################################## ### Supplementary functions rscores <- function (y, z, st = NULL, tau = 0, method=c("wilcoxon", "cs1", "cs2", "savage", "cs4", "sh", "U545", "U788")) { method = match.arg(method) stopifnot(length(tau) == 1) stopifnot(length(y) == length(z)) if (is.null(st)) st <- rep(1, length(y)) stopifnot(length(st) == length(y)) ust <- unique(st) nst <- length(ust) if (tau != 0) y <- y - z * tau sc <- rep(NA, length(y)) for (i in 1:nst) { who <- st == ust[i] yi <- y[who] ni <- length(yi) if (ni == 1) { sc[who] <- 0 } else if (ni >= 2) { sc[who] <- score(rank(yi), ni, method=method) } } sc } score <- function(j, Ns, method=c("W", "wilcoxon", "cs1", "cs2", "savage", "cs4", "sh", "U545", "U788"), a=5){ method = match.arg(method) sj <- switch(method, W = j, wilcoxon = j/(Ns+1), cs1 = sapply(j, function(j) prod( seq(j/(Ns+1), (j+a-2)/(Ns+1), by=1/(Ns+1)))), cs2 = (j/(Ns+1))^(a-1), savage = sapply(j, function(j) sum(1/(1-1:j/(Ns+1)))), cs4 = -log(1-j/(Ns+1)), sh = sapply(j, function(j) ifelse(j<a, 0, prod(seq((j-a+1)/(Ns+1),(j-1)/Ns+1), by=1/(Ns+1) )) ), U545 = sapply(j, function(j) ifelse(j<5, 0, (Ns*choose(Ns, 5)^(-1))*sum( choose((j-1), (4-1):(5-1))*choose((Ns-j), (5-4):(5-5)) ))), U788 = sapply(j, function(j) ifelse(j<8, 0, (Ns*choose(Ns, 8)^(-1))*sum( choose((j-1), (7-1):(8-1))*choose((Ns-j), (8-7):(8-8)) ))) ) sj }
72d12635eed75cc355cbcbc2164cedb0bd5db104
b1de6b75f59c9d003c922b8c796fb04115224bfc
/man/dice.Rd
c30dec3ae737856655887847ef4a0d5575bb1723
[]
no_license
muschellij2/msseg
4e282048d785907c3558ec1f1c9ae371c299e2e0
ed85b43b4395d87500d15c66fbc5738b0c516868
refs/heads/master
2020-05-21T20:57:52.329193
2019-01-29T15:34:34
2019-01-29T15:34:34
62,831,084
0
0
null
null
null
null
UTF-8
R
false
true
427
rd
dice.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/dice.R \name{dice} \alias{dice} \title{Calculate Dice Coefficient} \usage{ dice(prediction.obj) } \arguments{ \item{prediction.obj}{object of class \code{\link{prediction-class}} from \code{\link{prediction}}} } \value{ Object of class \code{\link{performance}} } \description{ Calculate Dice Coefficient/Similarity Index from a prediction object }
056dcecbc2f260c47714e7f9c85e1ec2ac7adf8b
592b4e685540c531e2d2130048de1cc3cdf6503b
/man/is.memo.Rd
6657667211f422622bf1860b26df4b9286478b46
[]
no_license
rwetherall/memofunc
1a202d176dc1de62751ce06c3c2a11a2ec835c76
0dd0ee722f6454f9025701df1065b55ed2c6b75a
refs/heads/master
2021-06-23T11:35:13.089942
2021-02-23T09:38:02
2021-02-23T09:38:02
178,967,249
0
0
null
2021-02-18T00:19:33
2019-04-02T00:18:49
R
UTF-8
R
false
true
339
rd
is.memo.Rd
% Generated by roxygen2: do not edit by hand % Please edit documentation in R/memo.R \name{is.memo} \alias{is.memo} \title{Is Memo} \usage{ is.memo(f) } \arguments{ \item{f}{function, memo or otherwise} } \value{ \code{TRUE} if memo function, \code{FALSE} otherwise } \description{ Checks whether the passed function is a memo function. }
80f7248a2b1aff5b8e38e348f60c9bce45def717
333d0bdae713f70f99ec2e4a13af83bd036e60e6
/Trade_report.R
10346f618c4f66d6695f42ae1a76d49f750ce84e
[ "Apache-2.0" ]
permissive
minhsphuc12/Work
a21a8f3204aa25e2f2edce8ed02b408a7f802576
aeae49dd4e6c4529b10bdac3ebd37b5d575fd451
refs/heads/master
2021-01-10T06:43:07.912223
2015-11-27T08:01:20
2015-11-27T08:01:20
46,967,025
0
0
null
null
null
null
UTF-8
R
false
false
30,895
r
Trade_report.R
#Load packages library(rio) library(data.table) # TRADE REPORT # Read Data data_pathT0=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201510.xlsx') data_pathT1=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201509.xlsx') data_pathT2=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201508.xlsx') data_pathT3=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201507.xlsx') data_pathT4=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201506.xlsx') data_pathT12=file.path(getwd(),'VNDIRECT/Trade_Report/Trade201410.xlsx') TradeT0=import(data_pathT0) TradeT1=import(data_pathT1) TradeT2=import(data_pathT2) TradeT3=import(data_pathT3) TradeT4=import(data_pathT4) TradeT12=import(data_pathT12) DATE_CHECK = "2015-10-30 UTC" # Convert Data Dates: NO NEED # Add ActiveStatus TradeT0$STATUS = ifelse (((TradeT0$GTGD==0)& (TradeT0$NAV_END==0))& (TradeT0$NAV_INIT==0), "CLOSE", "ACTIVE") TradeT1$STATUS = ifelse (((TradeT1$GTGD==0)& (TradeT1$NAV_END==0))& (TradeT1$NAV_INIT==0), "CLOSE", "ACTIVE") TradeT2$STATUS = ifelse (((TradeT2$GTGD==0)& (TradeT2$NAV_END==0))& (TradeT2$NAV_INIT==0), "CLOSE", "ACTIVE") TradeT3$STATUS = ifelse (((TradeT3$GTGD==0)& (TradeT3$NAV_END==0))& (TradeT3$NAV_INIT==0), "CLOSE", "ACTIVE") TradeT4$STATUS = ifelse (((TradeT4$GTGD==0)& (TradeT4$NAV_END==0))& (TradeT4$NAV_INIT==0), "CLOSE", "ACTIVE") TradeT12$STATUS = ifelse (((TradeT12$GTGD==0)& (TradeT12$NAV_END==0))& (TradeT12$NAV_INIT==0), "CLOSE", "ACTIVE") # Add Careby TradeT0$CAREBY = ifelse(substring(TradeT0$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT0$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT0$ACTYPE,1,1)==3,"VIP","OTHER"))) TradeT1$CAREBY = ifelse(substring(TradeT1$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT1$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT1$ACTYPE,1,1)==3,"VIP","OTHER"))) TradeT2$CAREBY = ifelse(substring(TradeT2$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT2$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT2$ACTYPE,1,1)==3,"VIP","OTHER"))) TradeT3$CAREBY = ifelse(substring(TradeT3$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT3$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT3$ACTYPE,1,1)==3,"VIP","OTHER"))) TradeT4$CAREBY = ifelse(substring(TradeT4$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT4$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT4$ACTYPE,1,1)==3,"VIP","OTHER"))) TradeT12$CAREBY = ifelse(substring(TradeT12$ACTYPE,1,1)==1,"MASS", ifelse(substring(TradeT12$ACTYPE,1,1)==2,"MG", ifelse(substring(TradeT12$ACTYPE,1,1)==3,"VIP","OTHER"))) # Add First Debt Check TradeT0$FIRST_DEBT = ifelse( ((as.numeric (difftime (as.Date(TradeT0$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT0$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0) , "YES","NO") TradeT1$FIRST_DEBT = ifelse(((as.numeric (difftime (as.Date(TradeT1$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT1$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0), "YES","NO") TradeT2$FIRST_DEBT = ifelse(((as.numeric (difftime (as.Date(TradeT2$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT2$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0), "YES","NO") TradeT3$FIRST_DEBT = ifelse(((as.numeric (difftime (as.Date(TradeT3$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT3$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0), "YES","NO") TradeT4$FIRST_DEBT = ifelse(((as.numeric (difftime (as.Date(TradeT4$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT4$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0), "YES","NO") TradeT12$FIRST_DEBT = ifelse(((as.numeric (difftime (as.Date(TradeT12$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) >-30) & ((as.numeric (difftime (as.Date(TradeT12$FIRST_DAY_OF_DEAL,"%d/%m/%Y"), as.Date(DATE_CHECK)), units = "days")) <= 0), "YES","NO") # Sum NAV/Trade Amount/Trade Fee/Interest (Careby) Total_NAV_MASS = c() Total_NAV_MASS[1] = sum (subset(TradeT0,CAREBY== "MASS")$NAV_END) Total_NAV_MASS[2] = sum (subset(TradeT1,CAREBY== "MASS")$NAV_END) Total_NAV_MASS[3] = sum (subset(TradeT2,CAREBY== "MASS")$NAV_END) Total_NAV_MASS[4] = sum (subset(TradeT12,CAREBY== "MASS")$NAV_END) Total_NAV_MG = c() Total_NAV_MG[1] = sum (subset(TradeT0,CAREBY== "MG")$NAV_END) Total_NAV_MG[2] = sum (subset(TradeT1,CAREBY== "MG")$NAV_END) Total_NAV_MG[3] = sum (subset(TradeT2,CAREBY== "MG")$NAV_END) Total_NAV_MG[4] = sum (subset(TradeT12,CAREBY== "MG")$NAV_END) Total_NAV_VIP = c() Total_NAV_VIP[1] = sum (subset(TradeT0,CAREBY== "VIP")$NAV_END) Total_NAV_VIP[2] = sum (subset(TradeT1,CAREBY== "VIP")$NAV_END) Total_NAV_VIP[3] = sum (subset(TradeT2,CAREBY== "VIP")$NAV_END) Total_NAV_VIP[4] = sum (subset(TradeT12,CAREBY== "VIP")$NAV_END) Total_TRADE_MASS = c() Total_TRADE_MASS[1] = sum (subset(TradeT0,CAREBY== "MASS")$GTGD) Total_TRADE_MASS[2] = sum (subset(TradeT1,CAREBY== "MASS")$GTGD) Total_TRADE_MASS[3] = sum (subset(TradeT2,CAREBY== "MASS")$GTGD) Total_TRADE_MASS[4] = sum (subset(TradeT12,CAREBY== "MASS")$GTGD) Total_TRADE_MG = c() Total_TRADE_MG[1] = sum (subset(TradeT0,CAREBY== "MG")$GTGD) Total_TRADE_MG[2] = sum (subset(TradeT1,CAREBY== "MG")$GTGD) Total_TRADE_MG[3] = sum (subset(TradeT2,CAREBY== "MG")$GTGD) Total_TRADE_MG[4] = sum (subset(TradeT12,CAREBY== "MG")$GTGD) Total_TRADE_VIP = c() Total_TRADE_VIP[1] = sum (subset(TradeT0,CAREBY== "VIP")$GTGD) Total_TRADE_VIP[2] = sum (subset(TradeT1,CAREBY== "VIP")$GTGD) Total_TRADE_VIP[3] = sum (subset(TradeT2,CAREBY== "VIP")$GTGD) Total_TRADE_VIP[4] = sum (subset(TradeT12,CAREBY== "VIP")$GTGD) Total_TRADE_FEE_MASS = c() Total_TRADE_FEE_MASS[1] = sum (subset(TradeT0,CAREBY== "MASS")$FEE) Total_TRADE_FEE_MASS[2] = sum (subset(TradeT1,CAREBY== "MASS")$FEE) Total_TRADE_FEE_MASS[3] = sum (subset(TradeT2,CAREBY== "MASS")$FEE) Total_TRADE_FEE_MASS[4] = sum (subset(TradeT12,CAREBY== "MASS")$FEE) Total_TRADE_FEE_MG = c() Total_TRADE_FEE_MG[1] = sum (subset(TradeT0,CAREBY== "MG")$FEE) Total_TRADE_FEE_MG[2] = sum (subset(TradeT1,CAREBY== "MG")$FEE) Total_TRADE_FEE_MG[3] = sum (subset(TradeT2,CAREBY== "MG")$FEE) Total_TRADE_FEE_MG[4] = sum (subset(TradeT12,CAREBY== "MG")$FEE) Total_TRADE_FEE_VIP = c() Total_TRADE_FEE_VIP[1] = sum (subset(TradeT0,CAREBY== "VIP")$FEE) Total_TRADE_FEE_VIP[2] = sum (subset(TradeT1,CAREBY== "VIP")$FEE) Total_TRADE_FEE_VIP[3] = sum (subset(TradeT2,CAREBY== "VIP")$FEE) Total_TRADE_FEE_VIP[4] = sum (subset(TradeT12,CAREBY== "VIP")$FEE) Total_INTEREST_MASS = c() Total_INTEREST_MASS[1] = sum (subset(TradeT0,CAREBY== "MASS")$INTEREST) Total_INTEREST_MASS[2] = sum (subset(TradeT1,CAREBY== "MASS")$INTEREST) Total_INTEREST_MASS[3] = sum (subset(TradeT2,CAREBY== "MASS")$INTEREST) Total_INTEREST_MASS[4] = sum (subset(TradeT12,CAREBY== "MASS")$INTEREST) Total_INTEREST_MG = c() Total_INTEREST_MG[1] = sum (subset(TradeT0,CAREBY== "MG")$INTEREST) Total_INTEREST_MG[2] = sum (subset(TradeT1,CAREBY== "MG")$INTEREST) Total_INTEREST_MG[3] = sum (subset(TradeT2,CAREBY== "MG")$INTEREST) Total_INTEREST_MG[4] = sum (subset(TradeT12,CAREBY== "MG")$INTEREST) Total_INTEREST_VIP = c() Total_INTEREST_VIP[1] = sum (subset(TradeT0,CAREBY== "VIP")$INTEREST) Total_INTEREST_VIP[2] = sum (subset(TradeT1,CAREBY== "VIP")$INTEREST) Total_INTEREST_VIP[3] = sum (subset(TradeT2,CAREBY== "VIP")$INTEREST) Total_INTEREST_VIP[4] = sum (subset(TradeT12,CAREBY== "VIP")$INTEREST) # Calculate median of trade fee (Careby) Median_FEE_MASS = c() Median_FEE_MASS [1] = median (subset(subset(TradeT0,CAREBY== "MASS"), FEE>0) $FEE) Median_FEE_MASS [2] = median (subset(subset(TradeT1,CAREBY== "MASS"), FEE>0) $FEE) Median_FEE_MASS [3] = median (subset(subset(TradeT2,CAREBY== "MASS"), FEE>0) $FEE) Median_FEE_MASS [4] = median (subset(subset(TradeT12,CAREBY== "MASS"), FEE>0) $FEE) Median_FEE_MG = c() Median_FEE_MG [1] = median (subset(subset(TradeT0,CAREBY== "MG"), FEE>0) $FEE) Median_FEE_MG [2] = median (subset(subset(TradeT1,CAREBY== "MG"), FEE>0) $FEE) Median_FEE_MG [3] = median (subset(subset(TradeT2,CAREBY== "MG"), FEE>0) $FEE) Median_FEE_MG [4] = median (subset(subset(TradeT12,CAREBY== "MG"), FEE>0) $FEE) Median_FEE_VIP = c() Median_FEE_VIP [1] = median (subset(subset(TradeT0,CAREBY== "VIP"), FEE>0) $FEE) Median_FEE_VIP [2] = median (subset(subset(TradeT1,CAREBY== "VIP"), FEE>0) $FEE) Median_FEE_VIP [3] = median (subset(subset(TradeT2,CAREBY== "VIP"), FEE>0) $FEE) Median_FEE_VIP [4] = median (subset(subset(TradeT12,CAREBY== "VIP"), FEE>0) $FEE) # Count Account/ with Trade/ with Debt / with first Debt (Careby) Account_ALL_MASS = c() Account_ALL_MASS [1] = sum (!is.na(subset(TradeT0,CAREBY== "MASS")$STATUS [subset(TradeT0,CAREBY== "MASS")$STATUS != "CLOSE"])) Account_ALL_MASS [2] = sum (!is.na(subset(TradeT1,CAREBY== "MASS")$STATUS [subset(TradeT1,CAREBY== "MASS")$STATUS != "CLOSE"])) Account_ALL_MASS [3] = sum (!is.na(subset(TradeT2,CAREBY== "MASS")$STATUS [subset(TradeT2,CAREBY== "MASS")$STATUS != "CLOSE"])) Account_ALL_MASS [4] = sum (!is.na(subset(TradeT12,CAREBY== "MASS")$STATUS [subset(TradeT12,CAREBY== "MASS")$STATUS != "CLOSE"])) Account_ALL_MG = c() Account_ALL_MG [1] = sum (!is.na(subset(TradeT0,CAREBY== "MG")$STATUS [subset(TradeT0,CAREBY== "MG")$STATUS != "CLOSE"])) Account_ALL_MG [2] = sum (!is.na(subset(TradeT1,CAREBY== "MG")$STATUS [subset(TradeT1,CAREBY== "MG")$STATUS != "CLOSE"])) Account_ALL_MG [3] = sum (!is.na(subset(TradeT2,CAREBY== "MG")$STATUS [subset(TradeT2,CAREBY== "MG")$STATUS != "CLOSE"])) Account_ALL_MG [4] = sum (!is.na(subset(TradeT12,CAREBY== "MG")$STATUS [subset(TradeT12,CAREBY== "MG")$STATUS != "CLOSE"])) Account_ALL_VIP = c() Account_ALL_VIP [1] = sum (!is.na(subset(TradeT0,CAREBY== "VIP")$STATUS [subset(TradeT0,CAREBY== "VIP")$STATUS != "CLOSE"])) Account_ALL_VIP [2] = sum (!is.na(subset(TradeT1,CAREBY== "VIP")$STATUS [subset(TradeT1,CAREBY== "VIP")$STATUS != "CLOSE"])) Account_ALL_VIP [3] = sum (!is.na(subset(TradeT2,CAREBY== "VIP")$STATUS [subset(TradeT2,CAREBY== "VIP")$STATUS != "CLOSE"])) Account_ALL_VIP [4] = sum (!is.na(subset(TradeT12,CAREBY== "VIP")$STATUS [subset(TradeT12,CAREBY== "VIP")$STATUS != "CLOSE"])) Account_TRADE_MASS = c() Account_TRADE_MASS [1] = sum (!is.na(subset(TradeT0,CAREBY== "MASS")$GTGD [subset(TradeT0,CAREBY== "MASS")$GTGD >0])) Account_TRADE_MASS [2] = sum (!is.na(subset(TradeT1,CAREBY== "MASS")$GTGD [subset(TradeT1,CAREBY== "MASS")$GTGD >0])) Account_TRADE_MASS [3] = sum (!is.na(subset(TradeT2,CAREBY== "MASS")$GTGD [subset(TradeT2,CAREBY== "MASS")$GTGD >0])) Account_TRADE_MASS [4] = sum (!is.na(subset(TradeT12,CAREBY== "MASS")$GTGD [subset(TradeT12,CAREBY== "MASS")$GTGD >0])) Account_TRADE_MG = c() Account_TRADE_MG [1] = sum (!is.na(subset(TradeT0,CAREBY== "MG")$GTGD [subset(TradeT0,CAREBY== "MG")$GTGD >0])) Account_TRADE_MG [2] = sum (!is.na(subset(TradeT1,CAREBY== "MG")$GTGD [subset(TradeT1,CAREBY== "MG")$GTGD >0])) Account_TRADE_MG [3] = sum (!is.na(subset(TradeT2,CAREBY== "MG")$GTGD [subset(TradeT2,CAREBY== "MG")$GTGD >0])) Account_TRADE_MG [4] = sum (!is.na(subset(TradeT12,CAREBY== "MG")$GTGD [subset(TradeT12,CAREBY== "MG")$GTGD >0])) Account_TRADE_VIP = c() Account_TRADE_VIP [1] = sum (!is.na(subset(TradeT0,CAREBY== "VIP")$GTGD [subset(TradeT0,CAREBY== "VIP")$GTGD >0])) Account_TRADE_VIP [2] = sum (!is.na(subset(TradeT1,CAREBY== "VIP")$GTGD [subset(TradeT1,CAREBY== "VIP")$GTGD >0])) Account_TRADE_VIP [3] = sum (!is.na(subset(TradeT2,CAREBY== "VIP")$GTGD [subset(TradeT2,CAREBY== "VIP")$GTGD >0])) Account_TRADE_VIP [4] = sum (!is.na(subset(TradeT12,CAREBY== "VIP")$GTGD [subset(TradeT12,CAREBY== "VIP")$GTGD >0])) Account_INTEREST_MASS = c() Account_INTEREST_MASS [1] = sum (!is.na(subset(TradeT0,CAREBY== "MASS")$INTEREST [subset(TradeT0,CAREBY== "MASS")$INTEREST >0])) Account_INTEREST_MASS [2] = sum (!is.na(subset(TradeT1,CAREBY== "MASS")$INTEREST [subset(TradeT1,CAREBY== "MASS")$INTEREST >0])) Account_INTEREST_MASS [3] = sum (!is.na(subset(TradeT2,CAREBY== "MASS")$INTEREST [subset(TradeT2,CAREBY== "MASS")$INTEREST >0])) Account_INTEREST_MASS [4] = sum (!is.na(subset(TradeT12,CAREBY== "MASS")$INTEREST [subset(TradeT12,CAREBY== "MASS")$INTEREST >0])) Account_INTEREST_MG = c() Account_INTEREST_MG [1] = sum (!is.na(subset(TradeT0,CAREBY== "MG")$INTEREST [subset(TradeT0,CAREBY== "MG")$INTEREST >0])) Account_INTEREST_MG [2] = sum (!is.na(subset(TradeT1,CAREBY== "MG")$INTEREST [subset(TradeT1,CAREBY== "MG")$INTEREST >0])) Account_INTEREST_MG [3] = sum (!is.na(subset(TradeT2,CAREBY== "MG")$INTEREST [subset(TradeT2,CAREBY== "MG")$INTEREST >0])) Account_INTEREST_MG [4] = sum (!is.na(subset(TradeT12,CAREBY== "MG")$INTEREST [subset(TradeT12,CAREBY== "MG")$INTEREST >0])) Account_INTEREST_VIP = c() Account_INTEREST_VIP [1] = sum (!is.na(subset(TradeT0,CAREBY== "VIP")$INTEREST [subset(TradeT0,CAREBY== "VIP")$INTEREST >0])) Account_INTEREST_VIP [2] = sum (!is.na(subset(TradeT1,CAREBY== "VIP")$INTEREST [subset(TradeT1,CAREBY== "VIP")$INTEREST >0])) Account_INTEREST_VIP [3] = sum (!is.na(subset(TradeT2,CAREBY== "VIP")$INTEREST [subset(TradeT2,CAREBY== "VIP")$INTEREST >0])) Account_INTEREST_VIP [4] = sum (!is.na(subset(TradeT12,CAREBY== "VIP")$INTEREST [subset(TradeT12,CAREBY== "VIP")$INTEREST >0])) Account_FIRSTDEBT_MASS = c() Account_FIRSTDEBT_MASS[1] = sum (!is.na(subset(TradeT0,CAREBY== "MASS")$FIRST_DEBT [subset(TradeT0,CAREBY== "MASS")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MASS[2] = sum (!is.na(subset(TradeT1,CAREBY== "MASS")$FIRST_DEBT [subset(TradeT1,CAREBY== "MASS")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MASS[3] = sum (!is.na(subset(TradeT2,CAREBY== "MASS")$FIRST_DEBT [subset(TradeT2,CAREBY== "MASS")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MASS[4] = sum (!is.na(subset(TradeT12,CAREBY== "MASS")$FIRST_DEBT [subset(TradeT12,CAREBY== "MASS")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MG = c() Account_FIRSTDEBT_MG[1] = sum (!is.na(subset(TradeT0,CAREBY== "MG")$FIRST_DEBT [subset(TradeT0,CAREBY== "MG")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MG[2] = sum (!is.na(subset(TradeT1,CAREBY== "MG")$FIRST_DEBT [subset(TradeT1,CAREBY== "MG")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MG[3] = sum (!is.na(subset(TradeT2,CAREBY== "MG")$FIRST_DEBT [subset(TradeT2,CAREBY== "MG")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_MG[4] = sum (!is.na(subset(TradeT12,CAREBY== "MG")$FIRST_DEBT [subset(TradeT12,CAREBY== "MG")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_VIP = c() Account_FIRSTDEBT_VIP[1] = sum (!is.na(subset(TradeT0,CAREBY== "VIP")$FIRST_DEBT [subset(TradeT0,CAREBY== "VIP")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_VIP[2] = sum (!is.na(subset(TradeT1,CAREBY== "VIP")$FIRST_DEBT [subset(TradeT1,CAREBY== "VIP")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_VIP[3] = sum (!is.na(subset(TradeT2,CAREBY== "VIP")$FIRST_DEBT [subset(TradeT2,CAREBY== "VIP")$FIRST_DEBT == "YES"])) Account_FIRSTDEBT_VIP[4] = sum (!is.na(subset(TradeT12,CAREBY== "VIP")$FIRST_DEBT [subset(TradeT12,CAREBY== "VIP")$FIRST_DEBT == "YES"])) # Calculate proportion Account with Trade/ with Debt / with first Debt (Careby) PERCENTAGE_TRADE_MASS = Account_TRADE_MASS / Account_ALL_MASS PERCENTAGE_TRADE_MG = Account_TRADE_MG / Account_ALL_MG PERCENTAGE_TRADE_VIP = Account_TRADE_VIP / Account_ALL_VIP PERCENTAGE_INTEREST_MASS = Account_INTEREST_MASS / Account_ALL_MASS PERCENTAGE_INTEREST_MG = Account_INTEREST_MG / Account_ALL_MG PERCENTAGE_INTEREST_VIP = Account_INTEREST_VIP / Account_ALL_VIP PERCENTAGE_FIRSTDEBT_MASS = Account_FIRSTDEBT_MASS / Account_ALL_MASS PERCENTAGE_FIRSTDEBT_MG = Account_FIRSTDEBT_MG / Account_ALL_MG PERCENTAGE_FIRSTDEBT_VIP = Account_FIRSTDEBT_VIP / Account_ALL_VIP # Calculate proportion Account full withdraw (Careby) mergeT0 = merge (TradeT1, TradeT2, by = "AFACCTNO") mergeT0 = merge (TradeT0, mergeT0, by = "AFACCTNO") mergeT0 $ CHURN = ifelse((mergeT0$NAV_END.y > 20*mergeT0$NAV_END) & ((mergeT0$NAV_END.y > 20*mergeT0$NAV_END.x) & (mergeT0$NAV_END.y >=500000)) ,"CHURN","RETAIN") ACCOUNT_CHURN_T0_MASS = length (subset(subset (mergeT0, CAREBY == "MASS"), CHURN == "CHURN")$CHURN) ACCOUNT_CHURN_T0_MG = length (subset(subset (mergeT0, CAREBY == "MG"), CHURN == "CHURN")$CHURN) ACCOUNT_CHURN_T0_VIP = length (subset(subset (mergeT0, CAREBY == "VIP"), CHURN == "CHURN")$CHURN) PERCENTAGE_CHURN_T0_MASS = ACCOUNT_CHURN_T0_MASS / Account_ALL_MASS [3] PERCENTAGE_CHURN_T0_MG = ACCOUNT_CHURN_T0_MASS / Account_ALL_MG [3] PERCENTAGE_CHURN_T0_VIP = ACCOUNT_CHURN_T0_MASS / Account_ALL_VIP [3] # Calculate 90% Percenticle NAV TOP_90PERCENTILE_MASS = c() TOP_90PERCENTILE_MASS [1] = quantile (subset(TradeT0, CAREBY == "MASS")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MASS [2] = quantile (subset(TradeT1, CAREBY == "MASS")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MASS [3] = quantile (subset(TradeT2, CAREBY == "MASS")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MASS [4] = quantile (subset(TradeT12, CAREBY == "MASS")$NAV_END, 0.9, na.rm = "TRUE") TOP_90PERCENTILE_MG = c() TOP_90PERCENTILE_MG [1] = quantile (subset(TradeT0, CAREBY == "MG")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MG [2] = quantile (subset(TradeT1, CAREBY == "MG")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MG [3] = quantile (subset(TradeT2, CAREBY == "MG")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_MG [4] = quantile (subset(TradeT12, CAREBY == "MG")$NAV_END, 0.9, na.rm = "TRUE") TOP_90PERCENTILE_VIP = c() TOP_90PERCENTILE_VIP [1] = quantile (subset(TradeT0, CAREBY == "VIP")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_VIP [2] = quantile (subset(TradeT1, CAREBY == "VIP")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_VIP [3] = quantile (subset(TradeT2, CAREBY == "VIP")$NAV_END, 0.9, na.rm = "TRUE" ) TOP_90PERCENTILE_VIP [4] = quantile (subset(TradeT12, CAREBY == "VIP")$NAV_END, 0.9, na.rm = "TRUE") # Calculate NAV/Trade Fee/Interest top 10% (Careby) TOP_NAV_MASS = c() TOP_NAV_MASS[1] = sum (subset (subset(TradeT0, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[1]) $NAV_END) TOP_NAV_MASS[2] = sum (subset (subset(TradeT1, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[2]) $NAV_END) TOP_NAV_MASS[3] = sum (subset (subset(TradeT2, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[3]) $NAV_END) TOP_NAV_MASS[4] = sum (subset (subset(TradeT12, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[4]) $NAV_END) TOP_FEE_MASS = c() TOP_FEE_MASS[1] = sum (subset (subset(TradeT0, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[1]) $FEE) TOP_FEE_MASS[2] = sum (subset (subset(TradeT1, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[2]) $FEE) TOP_FEE_MASS[3] = sum (subset (subset(TradeT2, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[3]) $FEE) TOP_FEE_MASS[4] = sum (subset (subset(TradeT12, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[4]) $FEE) TOP_INTEREST_MASS = c() TOP_INTEREST_MASS[1] = sum (subset (subset(TradeT0, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[1]) $INTEREST) TOP_INTEREST_MASS[2] = sum (subset (subset(TradeT1, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[2]) $INTEREST) TOP_INTEREST_MASS[3] = sum (subset (subset(TradeT2, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[3]) $INTEREST) TOP_INTEREST_MASS[4] = sum (subset (subset(TradeT12, CAREBY == "MASS"), NAV_END >= TOP_90PERCENTILE_MASS[4]) $INTEREST) TOP_NAV_MG = c() TOP_NAV_MG[1] = sum (subset (subset(TradeT0, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[1]) $NAV_END) TOP_NAV_MG[2] = sum (subset (subset(TradeT1, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[2]) $NAV_END) TOP_NAV_MG[3] = sum (subset (subset(TradeT2, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[3]) $NAV_END) TOP_NAV_MG[4] = sum (subset (subset(TradeT12, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[4]) $NAV_END) TOP_FEE_MG = c() TOP_FEE_MG[1] = sum (subset (subset(TradeT0, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[1]) $FEE) TOP_FEE_MG[2] = sum (subset (subset(TradeT1, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[2]) $FEE) TOP_FEE_MG[3] = sum (subset (subset(TradeT2, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[3]) $FEE) TOP_FEE_MG[4] = sum (subset (subset(TradeT12, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[4]) $FEE) TOP_INTEREST_MG = c() TOP_INTEREST_MG[1] = sum (subset (subset(TradeT0, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[1]) $INTEREST) TOP_INTEREST_MG[2] = sum (subset (subset(TradeT1, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[2]) $INTEREST) TOP_INTEREST_MG[3] = sum (subset (subset(TradeT2, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[3]) $INTEREST) TOP_INTEREST_MG[4] = sum (subset (subset(TradeT12, CAREBY == "MG"), NAV_END >= TOP_90PERCENTILE_MG[4]) $INTEREST) TOP_NAV_VIP = c() TOP_NAV_VIP[1] = sum (subset (subset(TradeT0, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[1]) $NAV_END) TOP_NAV_VIP[2] = sum (subset (subset(TradeT1, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[2]) $NAV_END) TOP_NAV_VIP[3] = sum (subset (subset(TradeT2, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[3]) $NAV_END) TOP_NAV_VIP[4] = sum (subset (subset(TradeT12, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[4]) $NAV_END) TOP_FEE_VIP = c() TOP_FEE_VIP[1] = sum (subset (subset(TradeT0, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[1]) $FEE) TOP_FEE_VIP[2] = sum (subset (subset(TradeT1, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[2]) $FEE) TOP_FEE_VIP[3] = sum (subset (subset(TradeT2, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[3]) $FEE) TOP_FEE_VIP[4] = sum (subset (subset(TradeT12, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[4]) $FEE) TOP_INTEREST_VIP = c() TOP_INTEREST_VIP[1] = sum (subset (subset(TradeT0, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[1]) $INTEREST) TOP_INTEREST_VIP[2] = sum (subset (subset(TradeT1, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[2]) $INTEREST) TOP_INTEREST_VIP[3] = sum (subset (subset(TradeT2, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[3]) $INTEREST) TOP_INTEREST_VIP[4] = sum (subset (subset(TradeT12, CAREBY == "VIP"), NAV_END >= TOP_90PERCENTILE_VIP[4]) $INTEREST) # Calculate proportion Trade Fee/ Interest top 10% (Careby) PERCENTAGE_TOP_NAV_MASS = TOP_NAV_MASS / Total_NAV_MASS PERCENTAGE_TOP_FEE_MASS = TOP_FEE_MASS / Total_TRADE_FEE_MASS PERCENTAGE_TOP_INTEREST_MASS = TOP_INTEREST_MASS / Total_INTEREST_MASS PERCENTAGE_TOP_NAV_MG = TOP_NAV_MG / Total_NAV_MG PERCENTAGE_TOP_FEE_MG = TOP_FEE_MG / Total_TRADE_FEE_MG PERCENTAGE_TOP_INTEREST_MG = TOP_INTEREST_MG / Total_INTEREST_MG PERCENTAGE_TOP_NAV_VIP = TOP_NAV_VIP / Total_NAV_VIP PERCENTAGE_TOP_FEE_VIP = TOP_FEE_VIP / Total_TRADE_FEE_VIP PERCENTAGE_TOP_INTEREST_VIP = TOP_INTEREST_VIP / Total_INTEREST_VIP # Calculate proportion Full Withdraw top 10% (Careby) TOP_CHURN_T0_MASS = length (subset ( subset (subset(mergeT0, CAREBY == "MASS"), CHURN == "CHURN"), NAV_END.y >= TOP_90PERCENTILE_MASS[3]) $CHURN) TOP_CHURN_T0_MG = length (subset ( subset (subset(mergeT0, CAREBY == "MG"), CHURN == "CHURN"), NAV_END.y >= TOP_90PERCENTILE_MG[3]) $CHURN) TOP_CHURN_T0_VIP = length (subset ( subset (subset(mergeT0, CAREBY == "VIP"), CHURN == "CHURN"), NAV_END.y >= TOP_90PERCENTILE_VIP[3]) $CHURN) PERCENTAGE_TOP_CHURN_T0_MASS = TOP_CHURN_T0_MASS / Account_ALL_MASS [3] PERCENTAGE_TOP_CHURN_T0_MG = TOP_CHURN_T0_MG / Account_ALL_MG[3] PERCENTAGE_TOP_CHURN_T0_VIP = TOP_CHURN_T0_VIP / Account_ALL_VIP[3] output = data.frame() output [1,1] = "" output [1,2] = "" output [1,3] = "" output [1,4] = "" output [1,5] = "" output [1,6] = "" colnames (output) = c('Indicators','CAREBY','T_0','T_1','T_2','T_12') output = rbind (output,c('Tong NAV','MASS',Total_NAV_MASS)) output = rbind (output,c('','MG',Total_NAV_MG)) output = rbind (output,c('','VIP',Total_NAV_VIP)) output = rbind (output,c('Tong so TK','MASS',Account_ALL_MASS)) output = rbind (output,c('','MG',Account_ALL_MG)) output = rbind (output,c('','VIP',Account_ALL_VIP)) output = rbind (output,c('Tong so TK co GD trong thang','MASS',Account_TRADE_MASS)) output = rbind (output,c('','MG',Account_TRADE_MG)) output = rbind (output,c('','VIP',Account_TRADE_VIP)) output = rbind (output,c('Ti le TK co GD trong thang','MASS',PERCENTAGE_TRADE_MASS)) output = rbind (output,c('','MG',PERCENTAGE_TRADE_MG)) output = rbind (output,c('','VIP',PERCENTAGE_TRADE_VIP)) output = rbind (output,c('Tong gia tri GD trong thang','MASS',Total_TRADE_MASS)) output = rbind (output,c('','MG',Total_TRADE_MG)) output = rbind (output,c('','VIP',Total_TRADE_VIP)) output = rbind (output,c('Tong phi GD trong thang','MASS',Total_TRADE_FEE_MASS)) output = rbind (output,c('','MG',Total_TRADE_FEE_MG)) output = rbind (output,c('','VIP',Total_TRADE_FEE_VIP)) output = rbind (output,c('Phi GD cua KH dai dien','MASS',Median_FEE_MASS)) output = rbind (output,c('','MG',Median_FEE_MG)) output = rbind (output,c('','VIP',Median_FEE_VIP)) output = rbind (output,c('Ti le su dung STPC','MASS',PERCENTAGE_INTEREST_MASS)) output = rbind (output,c('','MG',PERCENTAGE_INTEREST_MG)) output = rbind (output,c('','VIP',PERCENTAGE_INTEREST_VIP)) output = rbind (output,c('Ti le su dung SPTC lan dau','MASS',PERCENTAGE_FIRSTDEBT_MASS)) output = rbind (output,c('','MG',PERCENTAGE_FIRSTDEBT_MG)) output = rbind (output,c('','VIP',PERCENTAGE_FIRSTDEBT_VIP)) output = rbind (output,c('Doanh thu tu SPTC','MASS',Total_INTEREST_MASS)) output = rbind (output,c('','MG',Total_INTEREST_MG)) output = rbind (output,c('','VIP',Total_INTEREST_VIP)) output = rbind (output,c('Ti le KH rut trang NAV','MASS',PERCENTAGE_CHURN_T0_MASS,c("","",""))) output = rbind (output,c('','MG',PERCENTAGE_CHURN_T0_MG,c("","",""))) output = rbind (output,c('','VIP',PERCENTAGE_CHURN_T0_VIP,c("","",""))) output = rbind (output,c('Muc thap nhat top 10% NAV','MASS',TOP_90PERCENTILE_MASS)) output = rbind (output,c('','MG',TOP_90PERCENTILE_MG)) output = rbind (output,c('','VIP',TOP_90PERCENTILE_VIP)) output = rbind (output,c('Ti le NAV top 10% / Tong NAV','MASS',PERCENTAGE_TOP_NAV_MASS)) output = rbind (output,c('','MG',PERCENTAGE_TOP_NAV_MG)) output = rbind (output,c('','VIP',PERCENTAGE_TOP_NAV_VIP)) output = rbind (output,c('Tong phi GD top','MASS',TOP_FEE_MASS)) output = rbind (output,c('','MG',TOP_FEE_MG)) output = rbind (output,c('','VIP',TOP_FEE_VIP)) output = rbind (output,c('Ti le phi GD top','MASS',PERCENTAGE_TOP_FEE_MASS)) output = rbind (output,c('','MG',PERCENTAGE_TOP_FEE_MG)) output = rbind (output,c('','VIP',PERCENTAGE_TOP_FEE_VIP)) output = rbind (output,c('Doanh thu SPTC top','MASS',TOP_INTEREST_MASS)) output = rbind (output,c('','MG',TOP_INTEREST_MG)) output = rbind (output,c('','VIP',TOP_INTEREST_VIP)) output = rbind (output,c('Ti le doanh thu SPTC top','MASS',PERCENTAGE_TOP_INTEREST_MASS)) output = rbind (output,c('','MG',PERCENTAGE_TOP_INTEREST_MG)) output = rbind (output,c('','VIP',PERCENTAGE_TOP_INTEREST_VIP)) output = rbind (output,c('Ti le rut trang NAV top','MASS',PERCENTAGE_TOP_CHURN_T0_MASS,c("","",""))) output = rbind (output,c('','MG',PERCENTAGE_TOP_CHURN_T0_MG,c("","",""))) output = rbind (output,c('','VIP',PERCENTAGE_TOP_CHURN_T0_VIP,c("","",""))) #output = rbind (output,c('','',)) #WriteToFile write.table(output,file = 'sampletrade.csv',row.names = FALSE, sep = "") Threshold_Asset = 10^5 Threshold_Trade = 10^5 Threshold_MVC = 5*10^9 mergeT0 $ CUSTOMER = ifelse (mergeT0$STATUS == "ACTIVE", 'Y', 'N') mergeT0 $ ASSET = ifelse (mergeT0$NAV_END > Threshold_Asset, 'Y', 'N') mergeT0 $ TRADE = ifelse (mergeT0$GTGD > Threshold_Trade, 'Y', 'N') mergeT0 $ MVC = ifelse (mergeT0$GTGD > Threshold_MVC, 'Y', 'N') mergeT0 $ ENTER_CUSTOMER = ifelse (mergeT0$STATUS.x == "CLOSE" & mergeT0$STATUS == "ACTIVE" ,'Y','N') mergeT0 $ ENTER_ASSET = ifelse (mergeT0$NAV_END.x <= Threshold_Asset & mergeT0$NAV_END > Threshold_Asset ,'Y','N') mergeT0 $ ENTER_TRADE = ifelse (mergeT0$GTGD.x <= Threshold_Trade & mergeT0$GTGD > Threshold_Trade ,'Y','N') mergeT0 $ ENTER_MVC = ifelse (mergeT0$GTGD.x <= Threshold_MVC & mergeT0$GTGD > Threshold_MVC ,'Y','N') mergeT0 $ CHURN_CUSTOMER = ifelse (mergeT0$STATUS.x == "ACTIVE" & mergeT0$STATUS == "CLOSE" ,'Y','N') mergeT0 $ CHURN_ASSET = ifelse (mergeT0$NAV_END.x > Threshold_Asset & mergeT0$NAV_END <= Threshold_Asset ,'Y','N') mergeT0 $ CHURN_TRADE = ifelse (mergeT0$GTGD.x > Threshold_Trade & mergeT0$GTGD <= Threshold_Trade ,'Y','N') mergeT0 $ CHURN_MVC = ifelse (mergeT0$GTGD.x > Threshold_MVC & mergeT0$GTGD <= Threshold_MVC ,'Y','N') table (mergeT0$CAREBY, mergeT0$CUSTOMER) table (mergeT0$CAREBY, mergeT0$ASSET) table (mergeT0$CAREBY, mergeT0$TRADE) table (mergeT0$CAREBY, mergeT0$MVC) table (mergeT0$CAREBY, mergeT0$CHURN_CUSTOMER) table (mergeT0$CAREBY, mergeT0$CHURN_ASSET) table (mergeT0$CAREBY, mergeT0$CHURN_TRADE) table (mergeT0$CAREBY, mergeT0$CHURN_MVC) table (mergeT0$CAREBY, mergeT0$ENTER_CUSTOMER) table (mergeT0$CAREBY, mergeT0$ENTER_ASSET) table (mergeT0$CAREBY, mergeT0$ENTER_TRADE) table (mergeT0$CAREBY, mergeT0$ENTER_MVC) length(mergeT0$CAREBY) length(mergeT0$ASSET) nrow(subset(mergeT0,STATUS=="ACTIVE" )) CHURN_CUSTOMER = nrow(subset(subset(mergeT0,STATUS.x == "ACTIVE"), STATUS == "CLOSE")) CHURN_ASSET = nrow(subset(subset(mergeT0,NAV_END.x > Threshold_Asset), NAV_END <= Threshold_Asset)) CHURN_TRADE = nrow(subset(subset(mergeT0,GTGD.x > Threshold_Trade), GTGD <= Threshold_Trade)) CHURN_MVC = nrow(subset(subset(mergeT0,GTGD.x > Threshold_MVC), GTGD <= Threshold_MVC)) ENTER_CUSTOMER = nrow(subset(subset(mergeT0,STATUS.x == "CLOSE"), STATUS == "ACTIVE")) ENTER_ASSET = nrow(subset(subset(mergeT0,NAV_END.x <= Threshold_Asset), NAV_END > Threshold_Asset)) ENTER_TRADE = nrow(subset(subset(mergeT0,GTGD.x <= Threshold_Trade), GTGD > Threshold_Trade)) ENTER_MVC = nrow(subset(subset(mergeT0,GTGD.x <= Threshold_MVC), GTGD > Threshold_MVC)) GROUP_CUSTOMER = nrow(subset(mergeT0,STATUS == "ACTIVE")) GROUP_ASSET = nrow(subset(mergeT0,NAV_END >Threshold_Asset)) GROUP_TRADE = nrow(subset(mergeT0,GTGD >Threshold_Trade)) GROUP_MVC = nrow (subset(mergeT0,GTGD >Threshold_MVC))
21cfb311c19c4a8cb4a528239bb3047dda719b8e
ffdea92d4315e4363dd4ae673a1a6adf82a761b5
/data/genthat_extracted_code/PBD/tests/test-test_PBD.R
57546b5ed399db56b12b300bad4ff57b7a9c9dae
[]
no_license
surayaaramli/typeRrh
d257ac8905c49123f4ccd4e377ee3dfc84d1636c
66e6996f31961bc8b9aafe1a6a6098327b66bf71
refs/heads/master
2023-05-05T04:05:31.617869
2019-04-25T22:10:06
2019-04-25T22:10:06
null
0
0
null
null
null
null
UTF-8
R
false
false
552
r
test-test_PBD.R
context("test_PBD") test_that("PBD works", { pbd_sim(c(0.2,1,0.2,0.1,0.1),15) expect_equal(2 * 2, 4) res1 <- pbd_ML( brts = 1:10, initparsopt = c(4.62,4.34,0.03), exteq = 1, cond = 1) res2 <- pbd_ML( brts = 1:10, initparsopt = c(4.62,4.34,0.03), exteq = 1, cond = 2, n_l = 2, n_u = Inf) res2a <- pbd_ML( brts = 1:10, initparsopt = c(4.62,4.34,0.03), exteq = 1, cond = 2, n_l = 2, n_u = 1000) expect_equal(res1, res2) expect_equal(res2, res2a) })