Id
stringlengths
1
6
PostTypeId
stringclasses
7 values
AcceptedAnswerId
stringlengths
1
6
ParentId
stringlengths
1
6
Score
stringlengths
1
4
ViewCount
stringlengths
1
7
Body
stringlengths
0
38.7k
Title
stringlengths
15
150
ContentLicense
stringclasses
3 values
FavoriteCount
stringclasses
3 values
CreationDate
stringlengths
23
23
LastActivityDate
stringlengths
23
23
LastEditDate
stringlengths
23
23
LastEditorUserId
stringlengths
1
6
OwnerUserId
stringlengths
1
6
Tags
list
7009
1
7010
null
2
919
I have the pdf $$f(y ; \theta) = \frac{1}{\theta} \exp( \frac{-y}{\theta}), \ y > 0$$ and I'm supposed to determine if the following two estimators are unbiased or not: $ \hat \theta = nY_{min} $ and $ \hat \theta = \frac{1}{n}\sum_{i=1}^n Y_i $. I'm running into some problems because when I try to find the expected ...
Seeing if estimators are unbiased
CC BY-SA 2.5
null
2011-02-09T04:10:04.727
2011-02-09T23:10:24.540
2011-02-09T09:32:50.203
8
2387
[ "self-study", "unbiased-estimator" ]
7010
2
null
7009
2
null
Please see the following Wikipedia [article](http://en.wikipedia.org/wiki/Exponential_distribution) and it shows that both the integral exist. Also, if you look lower in the article, you will also see that both the estimators are unbiased. $Y_{min}$ is also exponentially distributed with mean parameter $\frac {\theta}...
null
CC BY-SA 2.5
null
2011-02-09T04:26:13.793
2011-02-09T09:33:03.997
2011-02-09T09:33:03.997
8
1307
null
7011
2
null
3531
4
null
WinBUGS has support for RJMCMC with an addon. I've used it for GLMs, including those with an intrinsic CAR component. Not R, obviously, but through R2WinBUGS you can patch them together.
null
CC BY-SA 2.5
null
2011-02-09T04:39:53.350
2011-02-09T04:39:53.350
null
null
26
null
7012
2
null
6990
1
null
This post explains it quite well: [http://blog.gribblelab.org/2009/03/09/repeated-measures-anova-using-r/](http://blog.gribblelab.org/2009/03/09/repeated-measures-anova-using-r/)
null
CC BY-SA 2.5
null
2011-02-09T06:40:13.363
2011-02-09T06:40:13.363
null
null
582
null
7013
2
null
841
4
null
The difficulty clearly comes because $X$ and $Y$ are corellated (I assume $(X,Y)$ is jointly gaussian, as Aniko) and you can't make a difference (as in @svadali's answer) or a ratio (as in Standard Fisher-Snedecor "F-test") because those would be of dependent $\chi^2$ distribution, and because you don't know what this ...
null
CC BY-SA 2.5
null
2011-02-09T08:45:33.313
2011-03-13T20:31:13.040
2011-03-13T20:31:13.040
223
223
null
7015
1
7017
null
2
5430
- Is linear regression only suitable for variables with a normal distribution? - If so, is there an alternative nonparametric test to test mediation or moderation?
Normal distribution necessary to assess moderating and mediating effects?
CC BY-SA 2.5
null
2011-02-09T11:01:45.140
2011-02-09T20:01:38.423
2011-02-09T11:10:17.940
183
null
[ "distributions", "mediation", "interaction", "multiple-regression" ]
7017
2
null
7015
6
null
For those not familiar with the language, moderation and mediation were both discussed in Barron and Kenny's influential article ([free pdf](http://www.public.asu.edu/~davidpm/classes/psy536/Baron.pdf)). ### Mediation With regards to mediation, bootstrapping is often used where normality does not seem like a reasona...
null
CC BY-SA 2.5
null
2011-02-09T12:17:35.667
2011-02-09T12:17:35.667
null
null
183
null
7019
1
null
null
1
435
I have a data set with a range of 0 to 65,000. The vast majority of data points (it is a huge sample) are concentrated between 0 and 1000. There is only one point that has 65,000. I want to plot this using a semi-logarithmic plot. However, I would like the graph to have around 50 points. If I use scales like 2,4,6...
Log-scale with concentrated data using integers
CC BY-SA 2.5
null
2011-02-09T13:16:41.170
2011-04-11T01:13:15.487
null
null
2405
[ "scales", "logarithm" ]
7020
1
7021
null
9
3625
Suppose I will be getting some samples from a binomial distribution. One way to model my prior knowledge is with a Beta distribution with parameters $\alpha$ and $\beta$. As I understand it, this is equivalent to having seen "heads" $\alpha$ times in $\alpha + \beta$ trials. As such, a nice shortcut to doing the ful...
Bayesian inference for multinomial distribution with asymmetric prior knowledge?
CC BY-SA 2.5
null
2011-02-09T14:22:32.753
2011-02-09T23:27:35.783
2011-02-09T23:27:35.783
2485
2485
[ "probability", "bayesian", "prior", "multinomial-distribution", "dirichlet-distribution" ]
7021
2
null
7020
3
null
You have framed your question very well. I think what you are looking for here is a case of hierarchical modeling. And you may want to model multiple layers of hierarchy (at the moment you only talk about priors). Having another layer of hyper-priors for the hyper--parameters lets you model the additional variabilities...
null
CC BY-SA 2.5
null
2011-02-09T14:51:50.390
2011-02-09T17:09:59.697
2011-02-09T17:09:59.697
1307
1307
null
7022
1
null
null
5
3882
Given a set of data (~5000 values) I'd like to draw random samples from the same distribution as the original data. The problem is there is no way to know for sure what distribution the original data comes from. It makes sense to use normal distribution in my case, although I'd like to be able to motivate that decisio...
Parameter estimation for normal distribution in Java
CC BY-SA 2.5
0
2011-02-09T16:50:00.400
2012-01-19T13:04:51.117
2011-02-09T17:13:52.090
null
3014
[ "estimation", "normal-distribution", "java" ]
7024
2
null
7019
2
null
Can you just use a scale comprised of powers of (1+r) for some small r, and round to the nearest integer? For example, in R, with r = 0.25: ``` > x <- unique(round(1.25^(0:50))) > x [1] 1 2 3 4 5 6 7 9 12 15 18 23 28 36 44 56 69 87 108 136 169 212 ...
null
CC BY-SA 2.5
null
2011-02-09T17:35:34.287
2011-02-09T17:35:34.287
null
null
2425
null
7026
2
null
7007
1
null
There are all kinds of analyses you can do. It really depends on what you want to know. Let me show you, with the example of gender: If you want to know wheter there are gender differences, you can apply a variance test (univariate or multivariate) on the likert scales with the demographic variable as independent facto...
null
CC BY-SA 2.5
null
2011-02-09T19:12:23.193
2011-02-09T19:12:23.193
null
null
1435
null
7029
1
7039
null
26
1295
I ran across this density the other day. Has someone given this a name? $f(x) = \log(1 + x^{-2}) / 2\pi$ The density is infinite at the origin and it also has fat tails. I saw it used as a prior distribution in a context where many observations were expected to be small, though large values were expected as well.
Does the distribution $\log(1 + x^{-2}) / 2\pi$ have a name?
CC BY-SA 2.5
null
2011-02-09T19:34:28.307
2011-02-12T05:07:24.860
2011-02-12T05:07:24.860
183
319
[ "distributions", "probability" ]
7030
2
null
7022
4
null
How big are the samples that you need? If substantially smaller than the 5000 points you have, say maximum 100 points or so, you could just take a random subset of your sample. Then you don't even need to assume normality - it's guaranteed to come from the distribution you want! Otherwise, it seems that the `org.apache...
null
CC BY-SA 2.5
null
2011-02-09T19:37:08.393
2011-02-09T19:37:08.393
null
null
2898
null
7031
2
null
7015
2
null
Many of us use linear regression in rough-and-ready fashion to learn about the relative importance of predictors, to assess the shape of relationships, and so on. But if one wants to make strict probabilistic inferences one needs to satisfy the set of standard assumptions entailed in such regression. The most importa...
null
CC BY-SA 2.5
null
2011-02-09T20:01:38.423
2011-02-09T20:01:38.423
null
null
2669
null
7032
1
7043
null
6
2040
Why is it that "missing data" and "outliers" can affect the performance of least square estimation?
Effect of missing data and outliers on least square estimation
CC BY-SA 2.5
null
2011-02-09T20:20:17.480
2012-07-24T10:41:12.550
2011-02-12T05:03:00.997
183
3125
[ "regression", "estimation", "least-squares" ]
7033
2
null
7029
5
null
Perhaps not. I could not find it in this fairly extensive list of distributions: [Leemis and McQuestion 2008 Univariate Distribution Relationships. American Statistician 62(1) 45:53](http://www.math.wm.edu/~leemis/2008amstat.pdf)
null
CC BY-SA 2.5
null
2011-02-09T23:08:02.450
2011-02-10T20:59:20.080
2011-02-10T20:59:20.080
449
2750
null
7034
2
null
7009
0
null
The fact that the sample mean is an unbiased estimator is obtained combining these two facts: 1. The sample mean is an unbiased estimator of the population mean 2. The population mean is equal to theta
null
CC BY-SA 2.5
null
2011-02-09T23:10:24.540
2011-02-09T23:10:24.540
null
null
null
null
7036
1
7041
null
6
4618
This question may have been asked before, but I couldn't find it. So, here goes. From about 3000 data points that can be characterized as "wins" or "losses" (binomial), it turns out that there are 52.8% dumb luck wins. This is my dependent variable. I also have some additional data that may help in predicting the a...
Number of trials required from a binomial distribution to get the desired odds
CC BY-SA 2.5
null
2011-02-10T01:24:08.580
2011-02-13T20:36:13.740
2011-02-13T20:36:13.740
2775
2775
[ "r", "binomial-distribution", "p-value" ]
7037
2
null
7032
4
null
If you're using R, try the following example. ``` library(tcltk) demo(tkcanvas) ``` Move the dots around to create all of the outliers you want. The regression will keep up with you.
null
CC BY-SA 2.5
null
2011-02-10T02:16:59.203
2011-02-10T02:16:59.203
null
null
2775
null
7038
2
null
7036
4
null
``` numtri[c(min(which(perwin <= 0.55)),max(which(perwin >= 0.55)))] ```
null
CC BY-SA 2.5
null
2011-02-10T02:20:51.027
2011-02-10T02:20:51.027
null
null
159
null
7039
2
null
7029
15
null
Indeed, even the first moment does not exist. The CDF of this distribution is given by $$F(x) = 1/2 + \left(\arctan(x) - x \log(\sin(\arctan(x)))\right)/\pi$$ for $x \ge 0$ and, by symmetry, $F(x) = 1 - F(|x|)$ for $x \lt 0$. Neither this nor any of the obvious transforms look familiar to me. (The fact that we can o...
null
CC BY-SA 2.5
null
2011-02-10T02:41:06.410
2011-02-10T02:41:06.410
null
null
919
null
7040
1
7042
null
4
14543
In Meta analysis, how to interpret the Egger’s linear regression method intercept (B0) 10.34631, 95% confidence interval (1.05905, 19.63357), with t=3.54535, df=3. The 1-tailed p-value (recommended) is 0.01911, and the 2-tailed p-value is 0.03822. I am a medical doctor. *Updated* The data are comparison of Regions 1 a...
Egger’s linear regression method intercept in meta analysis
CC BY-SA 2.5
null
2011-02-10T05:59:08.420
2011-02-10T15:35:02.927
2011-02-10T15:35:02.927
2956
2956
[ "meta-analysis", "funnel-plot", "publication-bias" ]
7041
2
null
7036
6
null
I wonder what it means to be "99% sure." The code seems to equate "dumb luck" with $p$ = 52.8% probability of wins. Let's imagine conducting $N$ trials, during which we observe $k$ wins. Suppose, for instance, $N$ = 1000 and you observe $k$ = 530 wins. That's greater than the expected number $p N$ = 528, but it's so...
null
CC BY-SA 2.5
null
2011-02-10T07:24:36.527
2011-02-10T14:25:46.007
2011-02-10T14:25:46.007
919
919
null
7042
2
null
7040
3
null
I suppose you are not interested in "hardcore" statistical explanation. So, more the intercept deviates from zero, the more pronounced the asymmetry. If the p-value of the intercept is 0.1 or smaller, the asymmetry is considered to be statistically significant. More [here](http://goo.gl/PlNhR).
null
CC BY-SA 2.5
null
2011-02-10T08:09:11.320
2011-02-10T08:09:11.320
null
null
609
null
7043
2
null
7032
6
null
I'm not sure about the "missing data", but I can give an answer on "outliers" This is basically due to the "unbounded" influence that a single observation can have in least squares (or at least in conventional least squares). A very, very simple example of least squares should show this. Suppose you only estimate an ...
null
CC BY-SA 2.5
null
2011-02-10T08:13:06.480
2011-02-10T08:13:06.480
null
null
2392
null
7045
1
10475
null
4
4318
Is there a package or library that can help me suggest a formula given the independent variables which will work well in glm, for example this formula can be something like x^2+log(y)+Z, it does not necessarily need to be the standard linear model x+y+z in order to explain a variable.
How to obtain in R a good formula for glm (general linear models) to predict a binomial variable?
CC BY-SA 2.5
null
2011-02-10T11:40:35.520
2012-07-31T19:27:54.183
2011-02-10T13:59:01.247
919
1808
[ "r", "logistic", "generalized-linear-model" ]
7046
2
null
7040
4
null
@Andrej has already given an answer, i.e there is evidence of funnel plot asymmetry. @DrWho, I would be interested in the reference that suggests using a one-tailed test. The following can give you an idea of the underlying logic of applying this regression model to test for publication bias: Most of these regressio...
null
CC BY-SA 2.5
null
2011-02-10T12:11:08.863
2011-02-10T12:11:08.863
null
null
307
null
7047
1
7067
null
2
1157
Hey guys. I got two images from video frames. They have a certain portion of overlap. After warping one of them, I'm currently trying to blend them together. In other words, I would like to stitch them together. But I don't know how to accomplish that. Can anybody please give me some help? Thank you! Let's say the ima...
Matlab image blending
CC BY-SA 2.5
null
2011-02-10T12:18:06.050
2011-02-10T21:15:49.847
null
null
3133
[ "matlab", "image-processing" ]
7048
1
7416
null
17
1385
While preparing for a talk I will give soon, I recently started digging into two major (Free) tools for interactive data visualization: [GGobi](http://www.ggobi.org/) and [mondrian](http://rosuda.org/mondrian/) - both offer a great range of capabilities (even if they're a bit buggy). I wish to ask for your help in art...
When is interactive data visualization useful to use?
CC BY-SA 2.5
null
2011-02-10T14:49:20.220
2013-02-22T00:04:37.230
2011-02-20T14:30:18.040
null
253
[ "data-visualization", "data-mining", "interactive-visualization" ]
7049
1
7076
null
13
15513
Can the poisson distribution be used to analyze continuous data as well as discrete data? I have a few data sets where response variables are continuous, but resemble a poisson distribution rather than a normal distribution. However, the poisson distribution is a discrete distribution and is usually concerned with numb...
Using poisson regression for continuous data?
CC BY-SA 3.0
null
2011-02-10T14:59:26.717
2015-07-14T19:46:48.197
2015-07-14T19:46:48.197
34826
3136
[ "distributions", "regression", "poisson-distribution", "continuous-data" ]
7050
2
null
7048
8
null
Dynamic linking of graphics is natural and effective for exploratory spatial data analysis, or [ESDA](http://www.ncgia.ucsb.edu/giscc/units/u128/u128_f.html). ESDA systems typically link one or more quantitative maps (such as [choropleth maps](http://en.wikipedia.org/wiki/Choropleth_map)) with tabular views and statis...
null
CC BY-SA 2.5
null
2011-02-10T15:13:17.287
2011-02-10T15:13:17.287
null
null
919
null
7051
2
null
7049
9
null
If you're talking about using a Poisson response in a generalized linear model, then yes, if you are willing to make the assumption that the variance of each observation is equal to its mean. If you don't want to do that, another alternative may be to transform the response (e.g. take logs).
null
CC BY-SA 2.5
null
2011-02-10T15:15:28.673
2011-02-10T15:15:28.673
null
null
495
null
7052
2
null
6225
2
null
Technically, no, a null hypothesis cannot be proven. For any fixed, finite sample size, there will always be some small but nonzero effect size for which your statistical test has virtually no power. More practically, though, you can prove that you're within some small epsilon of the null hypothesis, such that deviat...
null
CC BY-SA 2.5
null
2011-02-10T15:29:44.943
2011-02-10T15:29:44.943
null
null
1347
null
7053
1
7064
null
6
540
I’m reviewing an article, and can’t give details but here is the situation, and it’s got me puzzled Patients were divided into 4 categories (call them A B C and D), which were exhaustive and exclusive. Adjusted hazard ratios were computed for these four groups for all patients and for two subgroups of patients (call t...
Question about combining hazard ratios - Maybe Simpson's paradox?
CC BY-SA 3.0
null
2011-02-10T15:35:57.853
2017-03-06T18:43:56.290
2017-03-06T18:43:56.290
-1
686
[ "survival", "simpsons-paradox" ]
7054
1
7131
null
5
146
Can we do a meta-analysis of data of 3 regions. A particular disease was treated with the same treatment but implemented thoroughly in 2 regions and not so thoroughly in 1 region. How to proceed with analysis? What is the best way to analyse these data. The data are of a disease (cases and deaths) as follows: ``` ...
Meta-analysis of 3 regions' data for 5 years
CC BY-SA 2.5
null
2011-02-10T15:38:40.207
2011-02-12T16:39:30.410
2011-02-12T05:52:05.690
183
2956
[ "meta-analysis", "panel-data" ]
7056
1
7075
null
3
524
readHTMLTable seems pretty robust, but when I try to use it on this page, I get an error. Any ideas what it means or how I could get around it? The page's biggest table is nested inside another table... is that a problem? ``` > tables<-readHTMLTable(myURL, header=NA,a.data.frame=TRUE) Error in htmlParse(doc) : e...
Error creating parser in readHTMLTable
CC BY-SA 2.5
null
2011-02-10T16:11:00.267
2011-02-10T23:28:32.947
2011-02-10T17:53:56.070
null
1463
[ "r", "dataset" ]
7057
1
null
null
14
9313
I would like to get the coefficients for the LASSO problem $$||Y-X\beta||+\lambda ||\beta||_1.$$ The problem is that glmnet and lars functions give different answers. For the glmnet function I ask for the coefficients of $\lambda/||Y||$ instead of just $\lambda$, but I still get different answers. Is this expected? Wh...
GLMNET or LARS for computing LASSO solutions?
CC BY-SA 3.0
null
2011-02-10T16:23:48.187
2021-09-24T05:02:58.787
2016-08-08T02:43:39.467
805
null
[ "r", "machine-learning", "regression", "lasso", "regularization" ]
7058
1
7169
null
11
428
I have three features that I use to solve a classification problem. Originally, these features produced boolean values, so I could evaluate their redundancy by looking at how much the sets of positive and negative classifications overlap. Now I have extended the features to produce real values (scores) instead, and I w...
How to quantify redundancy of features?
CC BY-SA 2.5
null
2011-02-10T16:35:26.570
2011-02-13T22:04:12.390
2011-02-10T17:33:23.853
977
977
[ "correlation", "feature-selection" ]
7059
1
null
null
5
1320
I am managing many people entering data into a database. I have a log of user, date, time, table, and action that each person makes: ``` records <- data.frame(user = c('bob', 'bob', 'jane', 'jane', 'bob', 'bob', 'bob', 'jane', 'jane', 'bob'), date = c("2010-06-24", "2010-06-28", "2010-06-29", "201...
How can I effectively summarize and visualize time series of employee activities?
CC BY-SA 2.5
null
2011-02-10T18:02:16.093
2011-02-12T21:34:48.187
2011-02-11T08:08:51.310
2116
1381
[ "r", "time-series", "data-visualization" ]
7061
1
7078
null
3
500
Given two basketball players. John made 38/50 free throws. Mike made 80/100 free throws. What is probability that Mike is better at free throws than John?
Binomial Probability Question
CC BY-SA 2.5
null
2011-02-10T19:10:16.860
2011-02-11T04:02:45.067
null
null
3143
[ "probability", "binomial-distribution" ]
7062
2
null
7057
1
null
LASSO is non-unique in the case where multiple features have perfect collinearity. Here's a simple thought experiment to prove it. Let's say you have three random vectors $y$, $x_1$, $x_2$. You're trying to predict $y$ from $x_1$, $x_2$. Now assume $y$ = $x1$ = $x2$. An optimal LASSO solution would be $\beta_1 = ...
null
CC BY-SA 2.5
null
2011-02-10T19:19:24.763
2011-02-10T20:30:18.040
2011-02-10T20:30:18.040
1347
1347
null
7063
2
null
7053
2
null
Yes. It is certainly possible that this is due to something like Simpson's paradox. If the data looked like $$\begin{array}{rrrrrr} \textit{Organ}&\textit{Outcome}&A&B&C&D\\ \textrm{Lung}&\textrm{Bad}&371&2727&2374&418\\ \textrm{Lung}&\textrm{Good}&556&3199&2740&558\\ \textrm{Heart}&\textrm{Bad}&214&245&195&273\...
null
CC BY-SA 2.5
null
2011-02-10T20:13:28.700
2011-02-10T20:13:28.700
null
null
2958
null
7064
2
null
7053
5
null
Strictly, [Simpson's paradox](http://en.wikipedia.org/wiki/Simpson%27s_paradox) refers to a reversal in the direction of effect, which hasn't happened here as all the hazard ratios are above 1, so I'd refer to this by the more general term [confounding](http://en.wikipedia.org/wiki/Confounding). You can certainly have ...
null
CC BY-SA 2.5
null
2011-02-10T20:17:34.793
2011-02-10T20:17:34.793
null
null
449
null
7066
2
null
7061
1
null
I think what you want to do is compare the predictive distributions of both players. The predictive distribution describes the probability that Mike/John will make his next shot given the data (integrating out the parameters). Here is some Matlab code you can play with: ``` clear; clc; rng = linspace(0, 1, 100); % da...
null
CC BY-SA 2.5
null
2011-02-10T21:11:36.403
2011-02-10T21:11:36.403
null
null
1913
null
7067
2
null
7047
3
null
I don't know algorithms off the top of my head, but I would start by having a look at [Survey of image registration techniques](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.86.8364&rep=rep1&type=pdf)
null
CC BY-SA 2.5
null
2011-02-10T21:15:49.847
2011-02-10T21:15:49.847
null
null
1913
null
7068
2
null
7059
2
null
Below the code to plot the numbers of actions per week/per user: ``` load("records.Rdata") library(ggplot2) records$posdate <- as.POSIXlt(records$date,format="%Y-%m-%d") records$week <- as.numeric(format(records$posdate,"%W")) #changed from previous hack! numberActions <- by(records$action,records[,c("user","week")],fu...
null
CC BY-SA 2.5
null
2011-02-10T21:27:30.477
2011-02-11T09:08:07.167
2011-02-11T09:08:07.167
1443
1443
null
7069
1
null
null
3
327
Is it possible to correct for violating the assumption of independence for nonparametric tests? I have a categorical independent variable and a categorical and binary dependent variable, and each subject was exposed to multiple levels of treatment. I have done some preliminary analysis with binary logistic regression ...
Is it possible to correct for violating the assumption of independence for nonparametric tests?
CC BY-SA 2.5
null
2011-02-10T21:33:27.913
2011-02-10T22:48:44.833
null
null
null
[ "nonparametric", "non-independent" ]
7070
1
7071
null
27
115599
What is the best way of defining white noise process so it is intuitive and easy to understand?
What is a white noise process?
CC BY-SA 3.0
null
2011-02-10T22:13:25.977
2016-06-03T08:02:12.637
2012-08-20T06:18:15.407
2116
333
[ "time-series" ]
7071
2
null
7070
18
null
A white noise process is one with a mean zero and no correlation between its values at different times. See the ['white random process' section of Wikipedia's article on white noise](http://en.wikipedia.org/wiki/White_noise#White_random_process_.28white_noise.29).
null
CC BY-SA 2.5
null
2011-02-10T22:23:25.730
2011-02-10T22:23:25.730
null
null
449
null
7072
2
null
7069
2
null
You may be having trouble finding info on this because you're looking for it under 'nonparametric tests'. This wouldn't usually be included under the (rather confusing) 'nonparametric' heading. Logistic regression is sensible, but you'll need one of its extensions to deal with correlated outcomes, such as conditional ...
null
CC BY-SA 2.5
null
2011-02-10T22:30:01.883
2011-02-10T22:48:44.833
2011-02-10T22:48:44.833
449
449
null
7074
1
7083
null
10
14623
Is there a formal statistical test to test if process is a white noise?
Formal statistical test for whether a process is a white noise
CC BY-SA 2.5
null
2011-02-10T23:25:06.553
2018-12-02T08:21:33.307
2011-02-12T15:39:03.473
1036
333
[ "time-series", "white-noise" ]
7075
2
null
7056
2
null
The table in question seems to need use of a Javascript button to download. If you just want this particular data then save it as the offered untitle.txt or something else (it is basically tab-delimited) and then ``` tables <- read.delim("untitle.txt", skip=1) colnames(tables) <- c(colnames(tables)[-1],"jun...
null
CC BY-SA 2.5
null
2011-02-10T23:28:32.947
2011-02-10T23:28:32.947
null
null
2958
null
7076
2
null
7049
13
null
The key assumption of a generalized linear model that's relevant here is the relationship between the variance and mean of the response, given the values of the predictors. When you specify a Poisson distribution, what this implies is that you are assuming the conditional variance is equal to the conditional mean.* The...
null
CC BY-SA 2.5
null
2011-02-11T01:29:36.383
2011-02-11T01:29:36.383
null
null
1569
null
7077
2
null
7061
2
null
This is very much a problem in Bayesian inference. You appear to have taken the first step in realizing that the sample is not the same as the underlying probability distribution. Even though Mike had a higher mean in his sample, John might have a higher mean for his true talent distribution. When applying Bayesian ...
null
CC BY-SA 2.5
null
2011-02-11T03:47:17.390
2011-02-11T03:47:17.390
null
null
2485
null
7078
2
null
7061
2
null
Okay. I think I figured out the general answer: Given a sample of n/N made, the probability that the population success rate is greater than x is defined by the posterior probability distribution: > $\frac{\int_{x}^1 r^n(1-r)^{N-n}\ dr}{\int_{0}^1 r^n(1-r)^{N-n}\ dr}$ So for my example the chance that Mike's free th...
null
CC BY-SA 2.5
null
2011-02-11T03:55:13.450
2011-02-11T04:02:45.067
2011-02-11T04:02:45.067
3143
3143
null
7079
1
null
null
7
23409
- How do you carry out analysis of covariance using R? - How do you interpret the results? - A practical example will be highly appreciated.
Analysis of covariance in R
CC BY-SA 2.5
null
2011-02-11T06:27:08.743
2011-11-05T02:58:25.970
2011-02-12T05:49:02.013
183
3107
[ "r", "ancova" ]
7081
2
null
7070
12
null
I myself usually think of white noise as an iid sequence with zero mean. At different times values of the process are then independent of each other, which is much stronger requirement than correlation zero. What is the best with this definition that it works in any context. Side note. I only explained my intuition, t...
null
CC BY-SA 2.5
null
2011-02-11T07:35:13.443
2011-02-11T09:13:59.350
2011-02-11T09:13:59.350
2116
2116
null
7082
2
null
7079
7
null
Here is a [detailed presentation](http://www3.imperial.ac.uk/pls/portallive/docs/1/1171922.PDF)
null
CC BY-SA 2.5
null
2011-02-11T07:44:39.227
2011-02-11T07:44:39.227
null
null
339
null
7083
2
null
7074
14
null
In time-series analysis usually [Ljung-Box test](http://en.wikipedia.org/wiki/Ljung-Box_test) is used. Note though that it tests the correlations. If the correlations are zero, but variance varies, then the process is not white noise, but Ljung-Box test will fail to reject the null-hypothesis. Here is an example in R: ...
null
CC BY-SA 2.5
null
2011-02-11T07:47:54.650
2011-02-11T07:47:54.650
2017-04-13T12:44:56.303
-1
2116
null
7084
1
7543
null
3
2933
### Background: In connection with the question [here](https://stats.stackexchange.com/questions/7022/parameter-estimation-for-normal-distribution-in-java) I came upon a more interesting question. I believe the question is large and distinct enough to have it's own thread. Of course I might be mistaken, in that case...
Justifying normal approximation of experimental data
CC BY-SA 2.5
null
2011-02-11T09:09:36.410
2011-02-23T18:11:56.453
2017-04-13T12:44:20.840
-1
3014
[ "normality-assumption", "hypothesis-testing", "goodness-of-fit", "approximation" ]
7085
2
null
6225
2
null
There is a case where a proof is possible. Suppose you have a school and your null hypothesis is that the numbers of boys and of girls is equal. As the sample size increases, the uncertainty in the ratio of boys to girls tends to reduce, eventually reaching certainty (which is what I assume you mean by proof) when th...
null
CC BY-SA 2.5
null
2011-02-11T09:27:37.847
2011-02-11T09:27:37.847
null
null
2958
null
7086
2
null
7084
5
null
If your ultimate aim is, as you say, "to see if the results from experimental values show significant deviation from randomized data" then you'd be better to directly use the observed distribution by performing a [permutation (re-randomization) test](http://en.wikipedia.org/wiki/Resampling_%28statistics%29#Permutation_...
null
CC BY-SA 2.5
null
2011-02-11T09:31:33.927
2011-02-11T09:31:33.927
null
null
449
null
7088
1
7119
null
3
1261
I am doing some class work and I was wondering if there is a universally accepted format for analyzing and presenting the data in a Word document for data gathered from questionnaires.
Is there a standard format for presenting a data analysis report based on a questionnaire?
CC BY-SA 3.0
null
2011-02-11T11:41:17.083
2011-10-04T21:54:54.830
2011-10-04T21:54:54.830
183
3150
[ "survey", "reporting" ]
7089
1
null
null
22
18742
I need a formula for the probability of an event in a n-variate Bernoulli distribution $X\in\{0,1\}^n$ with given $P(X_i=1)=p_i$ probabilities for a single element and for pairs of elements $P(X_i=1 \wedge X_j=1)=p_{ij}$. Equivalently I could give mean and covariance of $X$. I already learned that there exist many $\{...
Probability formula for a multivariate-bernoulli distribution
CC BY-SA 2.5
null
2011-02-11T12:30:24.763
2022-01-18T05:14:29.933
2011-02-11T12:51:48.473
2116
null
[ "multivariate-analysis", "discrete-data" ]
7090
1
14308
null
4
400
When generating survival times to simulate Cox proportional hazards models, does it matter to generate them in days or in years? In theory, I guess it does not matter. But in practice? Is there a preference regarding computational issues? Thank you! Marco
Generating survival times in days or in years
CC BY-SA 2.5
null
2011-02-11T13:11:47.690
2011-08-17T16:51:46.080
2011-02-13T14:35:26.377
null
3019
[ "survival", "hazard", "units" ]
7091
2
null
7089
14
null
The random variable taking values in $\{0,1\}^n$ is a discrete random variable. Its distribution is fully described by probabilities $p_{\mathbf{i}}=P(X=\mathbf{i})$ with $\mathbf{i}\in\{0,1\}^n$. The probabilities $p_{i}$ and $p_{ij}$ you give are sums of $p_{\mathbf{i}}$ for certain indexes $\mathbf{i}$. Now it se...
null
CC BY-SA 3.0
null
2011-02-11T13:19:26.237
2017-12-02T19:35:33.767
2017-12-02T19:35:33.767
69834
2116
null
7093
2
null
5465
4
null
How will you count the number of sandal wood trees in Bangalore ?
null
CC BY-SA 2.5
null
2011-02-11T13:31:52.880
2011-02-11T13:31:52.880
null
null
null
null
7094
2
null
7089
1
null
I don't know what the resulting distribution is called, or if it even has a name, but it strikes me the obvious way to set this up is to think of the model you'd use to model a 2×2×2×…×2 table using a log-linear (Poisson regression) model. As you know the 1st-order interactions only, it's then natural to assume that al...
null
CC BY-SA 2.5
null
2011-02-11T13:39:43.833
2011-02-14T12:48:07.520
2011-02-14T12:48:07.520
449
449
null
7095
2
null
5465
3
null
We are running a customer service centre. We are getting 1 million calls per month. How do we reduce it to ten thousand ?
null
CC BY-SA 2.5
null
2011-02-11T13:40:00.073
2011-02-11T13:40:00.073
null
null
null
null
7096
1
null
null
1
529
Does anybody know whether R has a package/routine for estimating spline function with unknown number of knots?
Spline function with unknown knots
CC BY-SA 4.0
null
2011-02-11T13:48:51.977
2019-10-31T13:56:58.320
2019-10-31T13:56:58.320
92235
null
[ "r", "splines" ]
7097
2
null
7096
3
null
Have a look at the facilities for fitting smoothing splines in the `gam` package.
null
CC BY-SA 2.5
null
2011-02-11T13:56:45.177
2011-02-11T13:56:45.177
null
null
449
null
7098
2
null
7048
7
null
To me interactive visualization is useful only for my own exploration, or when working with a very hands-on client. When dealing with a final presentation, I prefer to choose the static graph that best makes my point. Otherwise clients can get totally distracted by the gee-whiz factor. The biggest benefit I get from i...
null
CC BY-SA 3.0
null
2011-02-11T14:21:54.300
2013-02-22T00:04:37.230
2013-02-22T00:04:37.230
-1
3155
null
7099
2
null
5465
16
null
Two questions I've been asked: 1) You fit a multiple regression to examine the effect of a particular variable a worker in another department is interested in. The variable comes back insignificant, but your co-worker says that this is impossible as it is known to have an effect. What would you say/do? 2) You have ...
null
CC BY-SA 2.5
null
2011-02-11T15:01:37.667
2011-02-11T15:01:37.667
null
null
2310
null
7100
1
null
null
7
3245
I am doing research and trying to use a questionnaire that is available in English in another language. - Is there a systematic approach for validating a questionnaire in another language? - What statistical test should I be undertake and why?
Validating an existing questionnaire into another language
CC BY-SA 2.5
null
2011-02-11T15:52:57.197
2015-11-09T12:13:56.343
2012-06-20T17:28:51.590
930
null
[ "correlation", "factor-analysis", "reliability", "psychometrics" ]
7101
1
null
null
4
2838
I am interested in applying Bayesian additive regression trees (BART) for classification analysis of gene expression data. I am relatively new to R (and Bioconductor packages) and I am unable to find some code or vignette that I can use to learn from. I will be thankful if someone can point me in a good direction.
Bayesian additive regression trees (BART) for classification analysis of gene expression data
CC BY-SA 2.5
null
2011-02-11T08:01:40.737
2022-12-17T03:34:55.783
2022-12-17T03:34:55.783
11852
4045
[ "r", "regression", "bayesian", "classification", "bart" ]
7102
1
7104
null
4
1748
I'm fitting a 4 parameter nonlinear regression model to multiple datasets, some of which fail to converge, however, the parameters output after a failure provide a fit that looks good, if not exceptional to my (and other's) eyes. I've explored convergence criteria and they do converge eventually but the visual fit is t...
Non-linear regression fails to converge, but fit appears good
CC BY-SA 3.0
null
2011-02-11T16:05:49.630
2011-12-27T15:21:40.740
2011-12-27T15:21:40.740
919
3158
[ "r", "curve-fitting", "nonlinear-regression" ]
7103
1
null
null
7
273
Example of problem: Part of our research team is working on providing operationally wind power forecast. Usually, since there are different time scalse that interest forecast user, a forecast is issued every 15 min (it has even happened that 5 seconds was requirer) for every 15 minutes ahead up to serveral days. Obviou...
Compression theory, practice, for time series with values in a space of distributions (say of a real random variable)
CC BY-SA 2.5
null
2011-02-11T17:29:15.263
2011-03-14T12:55:11.880
2011-02-12T07:23:26.243
223
223
[ "distributions", "time-series", "signal-processing", "quantiles", "compression" ]
7104
2
null
7102
7
null
I will assume the values of all the variables and constants are such that there won't be problems with obtaining square roots of negative numbers. Then $$\frac{\sqrt{c_1 x + c_2 \exp(x)^y}}{\sqrt{\exp(x)^y}} + c_3 =\sqrt{c_2 + c_1 x \exp(-y x)} + c_3.$$ When $y \gt 0$ then eventually, for sufficiently large $x$, $\fra...
null
CC BY-SA 2.5
null
2011-02-11T17:44:46.580
2011-02-11T21:06:12.897
2011-02-11T21:06:12.897
919
919
null
7105
2
null
3
4
null
This falls on the outer limits of 'statistical analysis', but [Eureqa](https://web.archive.org/web/20110201181704/http://ccsl.mae.cornell.edu/eureqa) is a very user friendly program for data-mining nonlinear relationships in data via genetic programming. Eureqa is not as general purpose, but it does what it does fairly...
null
CC BY-SA 4.0
null
2011-02-11T17:52:56.833
2022-11-27T23:12:34.187
2022-11-27T23:12:34.187
362671
795
null
7106
2
null
3
8
null
Colin Gillespie mentioned BUGS, but a better option for Gibbs Sampling, etc, is [JAGS](http://mcmc-jags.sourceforge.net/). If all you want to do is ARIMA, you can't beat [X12-ARIMA](https://web.archive.org/web/20120120205049/http://www.census.gov/srd/www/x12a/), which is a gold-standard in the field and open source. It...
null
CC BY-SA 4.0
null
2011-02-11T18:44:38.463
2022-11-27T23:22:26.533
2022-11-27T23:22:26.533
362671
1764
null
7107
2
null
7100
2
null
I found some good ideas in the short Sage book, Translating Questionnaires and Other Research Instruments, at [http://www.uk.sagepub.com/books/Book5861](http://www.uk.sagepub.com/books/Book5861) . I wouldn't call it the most systematic or the most entertaining read, but it was helpful and it's fairly up-to-date and in...
null
CC BY-SA 2.5
null
2011-02-11T18:46:41.513
2011-02-11T18:46:41.513
null
null
2669
null
7109
2
null
7101
2
null
I would suggest looking at the [BayesTree](http://cran.r-project.org/web/packages/BayesTree/index.html) package, from CRAN. I have no experience with it, so I cannot say if there are better option from there. Try looking at the [Machine Learning](http://cran.r-project.org/web/views/MachineLearning.html) Task View, or d...
null
CC BY-SA 2.5
null
2011-02-11T20:04:18.110
2011-02-11T20:04:18.110
null
null
930
null
7110
1
7147
null
39
30084
What is/are the difference(s) between a longitudinal design and a time series?
Difference between longitudinal design and time series
CC BY-SA 2.5
null
2011-02-11T22:51:58.117
2013-08-18T22:42:37.990
null
null
2956
[ "time-series", "panel-data" ]
7111
1
7114
null
18
14667
To perform principal component analysis (PCA), you have to subtract the means of each column from the data, compute the correlation coefficient matrix and then find the eigenvectors and eigenvalues. Well, rather, this is what I did to implement it in Python, except it only works with small matrices because the method t...
How to perform PCA for data of very high dimensionality?
CC BY-SA 3.0
null
2011-02-11T22:56:39.283
2021-08-09T13:53:56.283
2015-02-07T22:09:00.433
28666
null
[ "pca", "python" ]
7112
1
7118
null
45
47014
I have 2 simple questions about linear regression: - When is it advised to standardize the explanatory variables? - Once estimation is carried out with standardized values, how can one predict with new values (how one should standardize the new values)? Some references would be helpful.
When and how to use standardized explanatory variables in linear regression
CC BY-SA 4.0
null
2011-02-11T23:09:54.510
2018-06-26T16:38:23.270
2018-06-26T16:38:23.270
11887
1443
[ "regression", "predictive-models", "references", "standardization", "predictor" ]
7113
2
null
7110
13
null
A [time series](http://en.wikipedia.org/wiki/Time_series) is simple a sequence of data points spaced out over time, usually with regular time intervals. A [longitudinal design](http://en.wikipedia.org/wiki/Longitudinal_study) is rather more specific, keeping the same sample for each observation over time. An example o...
null
CC BY-SA 2.5
null
2011-02-12T00:20:17.573
2011-02-12T00:20:17.573
null
null
2958
null
7114
2
null
7111
11
null
The easiest way to do standard PCA is to center the columns of your data matrix (assuming the columns correspond to different variables) by subtracting the column means, and then perform an SVD. The left singular vectors, multiplied by the corresponding singular value, correspond to the (estimated) principal component...
null
CC BY-SA 3.0
null
2011-02-12T01:19:41.250
2015-02-07T22:09:44.477
2015-02-07T22:09:44.477
28666
1670
null
7115
1
7188
null
6
567
I'm wondering how far along the natural language processing is in determining the semantic distance between two excerpts of text. For instance, consider the following phrases - Early today, I got up and washed my car. - I cleaned my truck up this morning. - Bananas are an excellent source of potassium. Clearly (to...
Semantic distance between excerpts of text
CC BY-SA 2.5
null
2011-02-12T03:28:54.630
2022-06-20T14:19:06.153
2011-02-12T12:46:47.723
null
1026
[ "clustering", "classification", "text-mining" ]
7116
2
null
7111
11
null
It sounds like what you want is the NIPALS algorithm for performing PCA. It's a very popular algorithm among statisticians. It has many advantages: - Computationally less expensive than SVD or eigenvalue decomposition methods if only the first few components are required. - Has more modest storage requirements in gen...
null
CC BY-SA 2.5
null
2011-02-12T03:43:19.633
2011-02-12T03:43:19.633
null
null
2833
null
7117
2
null
7115
2
null
Check out the work by [Jones & Mewhort (2007)](https://doi.org/10.1037/0033-295X.114.1.1). This [more recent work](http://www.springerlink.com/content/qr50708142958x51/) may also be of interest, particularly their [online tool](http://www.casstools.org).
null
CC BY-SA 4.0
null
2011-02-12T03:50:03.520
2022-06-20T14:19:06.153
2022-06-20T14:19:06.153
361019
364
null
7118
2
null
7112
31
null
Although terminology is a contentious topic, I prefer to call "explanatory" variables, "predictor" variables. ### When to standardise the predictors: - A lot of software for performing multiple linear regression will provide standardised coefficients which are equivalent to unstandardised coefficients where you man...
null
CC BY-SA 2.5
null
2011-02-12T04:27:47.997
2011-02-12T11:13:03.750
2011-02-12T11:13:03.750
183
183
null
7119
2
null
7088
4
null
I'll do my best to interpret your question. ### Style Rules - Many journals and disciplines adopt a style guide (e.g., APA, Chicago, MLA, etc.). If one applies to you, then you are likely to have many relevant rules to guide you in the presentation of tables, figures, and much more. ### Table and Figure Design ...
null
CC BY-SA 2.5
null
2011-02-12T05:00:37.070
2011-02-12T05:00:37.070
null
null
183
null
7120
2
null
7088
2
null
Jeremy has offered many constructive suggestions. I'll add the point that > "If there were one approach that was clearly superior, then the Law of the Statistical Jungle would dictate that it would survive, and all of the other techniques would exist only as historical footnotes. The continued survival o...
null
CC BY-SA 2.5
null
2011-02-12T05:00:38.817
2011-02-12T05:18:18.587
2011-02-12T05:18:18.587
183
2669
null
7121
2
null
7032
1
null
An graphical example on outliers that requires no software and can be read in 2 minutes is [Wikipedia on Anscombe's quartet](http://en.wikipedia.org/wiki/Anscombe%27s_quartet)
null
CC BY-SA 3.0
null
2011-02-12T05:12:06.707
2012-07-24T10:41:12.550
2012-07-24T10:41:12.550
2795
2669
null
7122
2
null
6913
4
null
Just a quick thought: If you assume that each rating is drawn from a latent continuous variable, then you could define the median of this underlying continuous variable of interest as your value of interest, rather than the mean of this underlying distribution. Where the distribution is symmetric, then the mean and the...
null
CC BY-SA 2.5
null
2011-02-12T05:31:00.017
2011-02-12T05:31:00.017
null
null
183
null
7123
2
null
7110
25
null
If we think of designs made up of $n$ cases measured on $k$ occasions, then the following loose definition seems to me to be descriptive of the distinction: - longitudinal designs: high $n$, low $k$ - time series: low $n$, high $k$ Of course, this raises the question of what is high and what is low. Summarising my ...
null
CC BY-SA 2.5
null
2011-02-12T05:41:19.063
2011-02-13T11:51:25.673
2011-02-13T11:51:25.673
183
183
null
7124
2
null
7100
0
null
The procedure i have normally seen followed is to translate the questionnaire from english, and then have it back translated by someone else. If the two english translations match up, then you are good to go, otherwise repeat until they do.
null
CC BY-SA 2.5
null
2011-02-12T08:45:04.460
2011-02-12T08:45:04.460
null
null
656
null
7125
2
null
7100
7
null
I don't know what your questionnaire aims to assess. In Health-related Quality-of-Life studies, for example, there are a certain number of recommendations for translation issues that were discussed in the following papers (among others): - Marquis et al., Translating and evaluating questionnaires: Cultural issues for ...
null
CC BY-SA 2.5
null
2011-02-12T09:00:13.563
2011-02-12T09:00:13.563
2017-04-13T12:44:37.583
-1
930
null