question stringlengths 37 38.8k | group_id int64 0 74.5k |
|---|---|
<p>We have two linear fits, one for each data-set (unfortunately they include weights but I'm willing to ignore that if there's a nice analytic solution to this). Data-set <code>d1</code> and <code>d2</code> are different in the sense that <code>x</code> is the same variable but includes different discrete points (so <... | 70,677 |
<p>I wanted to study the distribution of BMI by age. So I randomly made age groups and did a chi square test. Is this appropriate?</p> | 13,524 |
<p>There is a data set given of course grades, with the variables: high school grade and gender, for which the gender one is a dummy that is equal to 1 if female.</p>
<p>I am to test the joint hypothesis that there are gender differences in grades in both the intercepts and slopes of the regression model. The coeffici... | 49,853 |
<p>What is a better measure of feature vector similarity, Euclidean distance or dot product/cosine similarity? I've read about cosine similarity being used with document vectors, but I've also seen distance being used in machine-learning/neural network methods.</p> | 70,678 |
<p>Given a data set of course grades, there is a female student dummy variable that is set to 1 if a student is female, and 0 if male. There is also a dummy for female instructors. </p>
<p>How would I perform the Hypothesis test using T-test with 5% significance, that the grades in classes taught by female instructors... | 49,853 |
<p>Supposing a Bayesian classifier with multivariate normal densities, how do I find the error rate of the classifier when we have two classes?</p>
<p>I am using this:</p>
<p>When dimension $d = 1$:</p>
<p>$$P(x | \mu , \sigma^2) = N(x, \mu, \sigma^2) = \frac{1}{\sqrt{2\pi \sigma^2}} e^{-\frac{(x-\mu)^2}{2\sigma^2}}... | 70,679 |
<p>I have a dataset with repeated measures at different speeds. I've binned the speed ranges into 0-0.5, 0.5-1.5, and 1.5-infinity. </p>
<pre><code>mydata`$`fact<−cut(mydata$section, breaks=c(0,0.5,1.5,Inf))
</code></pre>
<p>which gives the levels (0,0.5] (0.5,1.5] (1.5,Inf] </p>
<p>I first used the lme to test... | 70,680 |
<p>How to compute the overall standard error of a linear regression model using Python? Which library should I use? I am looking for something like <a href="http://stackoverflow.com/questions/11479064/multivariate-linear-regression-in-python">this</a>, however, I can't see how to get the overall standard error. Any ide... | 70,681 |
<p>I am trying to gain better understanding of the idea of <em>local outliers</em> (as discussed in this <a href="http://www.cse.ust.hk/~leichen/courses/msc-it5210/lectures/LOF_Example.pdf" rel="nofollow">pdf</a>) and how the function is implemented. Here are the key passages from the pdf: </p>
<blockquote>
<ul>
<... | 18,233 |
<p>I wonder if anyone can help me.</p>
<p>My data has 3 main variables: proportions at 2 time periods, and an additional predictor.
For example: </p>
<pre> Item Type Y(t) X Y(t+1)
1 .05 .2 .7
1 .07 .14 .06
1 0 .11 ... | 70,682 |
<p><a href="https://is.muni.cz/el/1456/jaro2009/PMAPEM/NotSortedYet/Campbell_Lo___MacKinlay_1997_The_Econometrics_of_Financial_Markets.pdf" rel="nofollow">Campbell (1997)</a> showed that there are three types of random walk hypotheses classified by the level of restrictions on $\epsilon_t$:</p>
<ol>
<li>RW1 (random wa... | 70,683 |
<p>I have a time series, let's call it $y(t)$. Augmented Dickey-Fuller and Phillips-Perron tests indicate no unit root but a deterministic trend. The series is strongly autocorrelated. I estimated the following short-run dynamic model: </p>
<p>$$y(t) = \alpha + \beta\cdot y(t-1) + \gamma\cdot t + \varepsilon_t$$</p>
... | 70,684 |
<p>I'm modeling longitudinal patient level health records data which will have the following structure:</p>
<p>\begin{equation*}
\text{Prob}(Y_{it}=1 \text{ at time } t \text{ and member } i)=\beta_{i0} +\beta_{i1}x_{i,t-1} + \beta_{i2}x_{i,t-2} + ... + \beta_{ik}x_{i,t-k} + \text{additional covariates for patient age... | 70,685 |
<p>I've been having some trouble with this story problem. Any help you could give me would be really appreciated. </p>
<p>A store manager monitors the store's temperature by taking 4 independent temperature readings at 4 locations where, if the system is working correctly, the temp. is normally distributed with a me... | 70,686 |
<p>I have conducted an experiment in which a participant completes memory tasks (3 levels, within) while exploring a virtual environment (2 levels, within). The performance on these memory tasks is considered a binomial dependent variable. In fact there are two learning periods, each with its own memory task. After exp... | 70,687 |
<p>I'm trying to figure out how to present results (e.g. precision, recall, accuracy, ROC curve) for a class project.</p>
<p>I have a multi-class classification problem(specific classification algorithm not really important) of classes <strong>A, B, C and D</strong>.</p>
<p>The classification algorithm was <strong>tr... | 49,854 |
<p>I performed a survey using a Likert 1 to 5 scale (totally agree/agree/neutral/ disagree/totally disagree) on 12 questions which are split into 3 statements which the respondent places a value of between 1 to 5 dependent on how much they agree or disagree - there are 36 statements in total.</p>
<p>Respondents:</p>
... | 49,578 |
<p>I am facing a situation where I need to assess the predictive ability of a single (regression) model. I do not need to compare its performance with some other model, I just need to show that it predicts well or not. I have the Mean Square Error and Mean Absolute error for both on the training set and a separate test... | 37,698 |
<p>I am a student and I have a project to do but I have some questions about ADF test. I have 52 observations of quarterly data. I have done the ADF test in my variables in the levels. I tested with intercept, intercept and trend and also using lags.</p>
<ol>
<li>The test with inercept and the one with intercept and l... | 3,190 |
<p>I need to run a cox-regression on a huge data set (millions) with only a few unique rows (hundreds) and regressing on categorical variable <code>group</code>. I could use just <code>Surv(t, c) ~ group</code>, but this would take a very long time. I was thinking about including only unique rows together with their co... | 70,688 |
<p>I'm examining a part of my dataset containing 46840 double values ranging from 1 to 1690 grouped in two groups. In order to analyze the differences between these groups I started by examining the distribution of the values in order to pick the right test.</p>
<p>Following a guide on testing for normality, I did a q... | 70,689 |
<p>suppose $X_1, X_2$ are independent random variable of $N(0,1)$.
suppose $$U=X_1+X_2, V=X_1^2+X_2^2$$ how can I calculate joint Moment generating function of $U,V$</p> | 70,690 |
<p>I'm trying to understand matrix notation, and working with vectors and matrices. </p>
<p><strong>Right now I'd like to understand how the vector of coefficient estimates $\hat{\beta}$ in multiple regression is computed.</strong></p>
<p>The basic equation seems to be </p>
<p>$$
\frac{d}{d\boldsymbol{\beta}} (\bo... | 37,809 |
<p>I would imagine this has been asked before but I can't find it so feel free to link if you can!</p>
<p>Lets say I have a sample size of 100 and an unknown population size. There's a great deal I can learn from this sample about the total population, for example the mean of the sample is likely to be close to the me... | 70,691 |
<p>I am looking for a better term for what I call '<em>uniformification</em>', where I change data to make it more close to uniformly distributed.</p>
<p>I am doing a project in which I try to make the output of a neural network uniformly distributed over the output space. Let's call the neural network $f(x | \theta)$... | 70,692 |
<p>I am wondering whether there is another probability density function known to literature which is similar to the multivariate normal distribution in the respect that the pdf incorporates the covariances between all variables, it has the covariance matrix as a parameter or is in some way defined by other dependencies... | 31,761 |
<p>I will very grateful with some help on the following problem:
I need to forecast probability of default for portfolio of retail loans, depending on several factors, that can be divided into three groups: first group – is initial conditions during the perdio, when these loans were issued (these factors are fixed in t... | 70,693 |
<p>I have a hypothetical data given below that consists of 11 pairs of points (xi, yi ), to which the simple linear regression mean function $\mathbb E(y|x) = β_0 + β_1x$ is fit.:</p>
<pre><code> X Y
10 8.04
8 6.95
13 7.58
9 8.81
11 8.33
14 9.96
6 7.24
4 4.26
12 10.84
7 ... | 31,763 |
<p>I would like to minimise the function $l(\theta|Y)$.</p>
<p>Given the Newton's method below</p>
<p>$$\theta^{(t+1)} = \theta^{(t)} - \left[l''(\theta\;|\;Y)\right]^{-1} l'(\theta^{(t)}\; | Y)\quad t = 0,1,...$$</p>
<p>where $l''(\theta^{(t)} | Y)$ is the observed Fisher Information matrix. </p>
<p>The observed f... | 49,855 |
<p>Very basic question here, so bear with me... </p>
<p>I have a data set with 241 patients with 16 variables plus diagnosis (malignant vs benign). There are 3 previously published logistic regression formulas in the field:</p>
<ol>
<li>modelS ~ A + B + C + D + E + F </li>
<li>modelH ~ modelS + G </li>
<li>modelG ~ A... | 70,694 |
<p>I have a 2x2x2 repeated measures design and am trying to decompose a 3-way interaction.
As all my factors have 2 levels, I always reach the point that I am doing pairwise comparisons for the simple effects. I have not managed to find a clear answer as to whether I should correct my p.values acquired from these depe... | 70,695 |
<p>Next week we have an intern from a local school in the house. The concept behind his short internship is to get an idea how the real world works and what certain jobs deal with, how the daily work looks like etc.</p>
<p>Now I wondered, what one can tell/show/demonstrate such a young kid about Statistics and Machine... | 70,696 |
<p>I am currently trying to fit distributions to some heavy tailed data-set (see the data set below) and have a hard time producing good results:</p>
<pre><code>v1 <- c(11.87,14.04,11.86,179.88,13.09,14.68,13.54,32.25,54.52,66.28,15.16,39.83,89.81,116.87,298.94,427,6249.42,3334.9,4503.93,4933.9,15.72,12.33,13.64,15... | 6,084 |
<p>Suppose you have two (or more) models that you want to ensemble together. However, some of the models are trained specifically on very specialized subsets of your data. If you do the stacking with a simple linear regression where the independent variables are the model outputs, how should you represent that output o... | 70,697 |
<p>For a given data matrix $A$, it seems like $A^TA$ plays an important role in statistics. For example, it is an important part of the analytical solution of ordinary least squares. Or, for PCA, its eigenvectors are the principal components of the data.</p>
<p>I understand how to calculate $A^TA$, but I was wondering... | 49,685 |
<p>"Big data" is everywhere in the media. Everybody says that "big data" is the big thing for 2012, e.g. <a href="http://www.kdnuggets.com/2012/01/hottest-analytics-data-mining-topics-2012.html">KDNuggets poll on hot topics for 2012</a>. However, I have deep concerns here. With big data, everybody seems to be happy jus... | 31,771 |
<p>I have failure data for an experience over $T$ years: at the beginning of each year I have $n_t$ subjects and $d_t$ of these subjects experience a "failure" at the end of the year. </p>
<p>Now if I assume that the trials are iid Binomial and the probability of failure $p$ is homogenous (i.e. constant over time), ... | 31,772 |
<p>I was wondering if anyone knows of any really productive ways to graph/map/plot/visualize covariance matrix data. I am currently using <code>levelplot()</code> in <code>R</code>. It's very basic, and does the job, but we're looking for something better. We're also using <code>R</code> Shiny so that there is more int... | 70,698 |
<p>My model is as follows:
$$
\begin{aligned}
y&=a_0 + a_1x_1 + a_2x_2 \\
x_2 &= b_0 + b_1x_1 + b_2z
\end{aligned}
$$</p>
<p>I'm only interested in the effect of $x_2$ on $y$. More precisely, I want to see the combined effect of: </p>
<ol>
<li>$x_2$ directly, and </li>
<li>$x_2$ implicitly (through $x_1$)</... | 70,699 |
<p>Does anyone know a way to calculate the effect size of a Fisher's exact test? Does that even make sense? </p> | 70,700 |
<p>I have to set up an example of how to do an MDL, and the professor wants to know how many times I'd run the samples. My previous class used as little as 8, but I'm not sure if thats correct or if I should say 20 instead.</p> | 31,777 |
<p>I have conducted a study in which I have gathered 3 sentences per participant. These sentences were then classified and the words per sentence were counted. I want to test if there are significant differences in the word count means for the different categories of the classification. </p>
<p>However, because the se... | 70,701 |
<p>I am trying to improve my data science knowledge by solving problems available on the internet. </p>
<p>I am currently using the R package <code>randomForest</code> to classify the data. By using all of the features available in the training set I get an AUC of about 87% and when I use my model on the test data I g... | 70,702 |
<blockquote>
<p><strong>Possible Duplicate:</strong><br>
<a href="http://stats.stackexchange.com/questions/26465/machine-learning-linear-classifier">Machine Learning: Linear classifier</a> </p>
</blockquote>
<p>Suppose we have 2 kernel functions $K_1(x,y)$ and $K_2(x,y)$. We know, that the dataset ($(x_1,y_1),\... | 49,856 |
<p>I'm trying to derive covariance from "embedded" random variable (r.v.). Say we have two Gaussian r.v.'s</p>
<p>$$ X_1 \sim N(\mu_1, \sigma^2_1)$$
$$ X_2 \sim N(\mu_2, \sigma^2_2)$$</p>
<p>where $\mu_1$ and $\mu_2$ are also Gaussian r.v,</p>
<p>$$\mu_1 \sim N(\theta_1, \delta^2_1)$$
$$\mu_2 \sim N(\theta_2, \delta... | 31,782 |
<p>I have data of an entire population (N=27). I would like to find out which variables (of many) have the greatest effect on one certain dependent variable, and how much of its variance can they explain. The possible independent variables could be correlating as well (some of them certainly do). They are all scale mea... | 70,703 |
<p>I just barely grasped why large-sample confidence limits that assume approximately normal distributions are not the best approach when relative frequencies approach either one or zero - so that in these instances methods according to Jeffrey or Clopper/Pearson are preferable, and can be calculated with relative ease... | 70,704 |
<p>Say we have a sample $X_{1},...,X_{n}$ from a log-normal distribution with parameters $\mu$ and $\sigma^{2}$. That is, $\ln(X)$~$N(\mu,\sigma^{2})$. Let $T_{n},Z_{n}$ denote the MLE's for $\mathbb{E}(X)$ and $\mathbb{D}^{2}(X)$ respectively (based on the sample of size $n$). The question is: What are the asymptotic ... | 445 |
<p>In testing a null vs an alternative, if we don't reject the null (e.g. the p-value is very big, bigger than the significance level), what is our conclusion?</p>
<p>Can we say that we accept the null, can we? I remember I heard that not rejecting the null doesn't mean accepting the null, but I might be wrong.</p>
<... | 49,424 |
<p>In Bayesian inference, we usually sample from the posterior $f(\theta_1,\theta_2|-)$ via MCMC to compute point estimates for the parameters of interest. I am investigating an alternate form of arriving at a point estimate. Sometimes it happens that drawing from the full posterior, $f(\theta_1,\theta_2|-)$, is diffic... | 31,785 |
<p>In a relatively simple setting, I try to fit a simple linear regression model. I want to assess how good this model for prediction. basically I try to do something like K-fold cross validation or jackknifing (that is - fit the model on part of the data, and predict the rest to see what happens). However, I seek a si... | 70,705 |
<p>I've tried to solve this but I have no idea; I tried it several times but I still didn't manage to get a correct solution.</p>
<p>For a sample of 10 students, the following bivariate data represents the distance and the duration of their travel to school.
$$\begin{array}{|c|c|} \hline
\mathbb x & 1 & 3 &am... | 31,786 |
<p>What is the best way to show a relationship between:</p>
<ul>
<li>continuous and discrete variable,</li>
<li>two discrete variables ?</li>
</ul>
<p>So far I have used scatter plots to look at the relationship between continuous variables. However in case of discrete variables data points are cumulated at certain i... | 70,706 |
<p>I have noticed that when I try to find the critical values for the Mann-Whitney U using R, the values are always 1+critical value. For example, for $\alpha=.05, n = 10, m = 5$, the (two-tailed) critical value is 8, while for $\alpha=.05, n=12, m=8$, the (two-tailed) critical value is 22 (check the <a href="http://ww... | 31,787 |
<p>In the following example I have a data frame which consists of a time series of water temperature measurements recorded at 5 depths in the ocean where each value in <code>Temp</code> corresponds to the date in <code>DateTime</code> and the depth in <code>Depth</code>. </p>
<pre><code>set.seed(1)
Temp <- rnorm(43... | 37,557 |
<p>I am researching Sentiment Analysis over social media, particularly classifying online texts such as blog posts as positive, negative or neutral. </p>
<p>Most of the approaches I have found for sentiment analysis are supervised (they need labeled data to train a classifier). However, I have also found a couple of p... | 70,707 |
<p>Suppose I want to test whether or not code patches created on weekends have a greater bug rate than those created on weekdays. (We might guess that this is so because people who are at work on weekends are more hurried etc.)</p>
<p>If we follow the standard and model bug creation as a Poisson process, this means (I... | 70,708 |
<p>I have heard that having more than one independent variable is useful when predicting human behaviour since it can be influenced by a combination of several factors!</p>
<p>Forgetting about the limitations like the number of observations, is there any way to decide how many independent variables to include in a mod... | 70,709 |
<p>I'm having trouble understanding the curse of dimensionality. Specifically, I came across it while doing the <code>scikit-learn</code> tutorial in python. Can someone please explain the below in a simpler manner? Sorry I have been trying to understand for the longest time and can't understand how they came up with t... | 70,710 |
<p>I came across the term 'teleoanalysis' in relation to medical trials. There is no Wikipedia article on it yet, but Google search has a few links to scientific papers and their critiques.</p>
<p>As far as I understand, one of the main concepts of teleoanalysis is: if A causes B, and B causes C, then we can conclude ... | 70,711 |
<p>I'm fitting a random effects model with <code>glmer</code> to some business data. The aim is to analyse sales performance by distributor, taking into account regional variation. I have the following variables:</p>
<ul>
<li><code>distcode</code>: distributor ID, with about 800 levels</li>
<li><code>region</code>: to... | 70,712 |
<p>I'm rebutting a paper in which the author has done linear regression in two steps:</p>
<ol>
<li>Regression against a single predictor</li>
<li>Regression of the residuals from the first model against a second predictor</li>
</ol>
<p>The problem with this approach is that if the two predictors are correlated (they ... | 70,713 |
<p>Let $X_1, \ldots, X_n$ be i.i.d. exponentially distributed random variables with density</p>
<p>$$\eqalign{\theta^{-1} e^{-x/\theta}, &x \ge 0 \\ 0, &x \lt 0} $$</p>
<p>and let $Y_i = X_{(i)}$ denote the order statistics such that $Y_1 \leq \cdots \leq Y_n$.</p>
<p>How to show that
$$
2\frac{\left(\s... | 70,714 |
<p>I have done an ordinal regression using <code>lrm</code> from the <code>rms</code> package.</p>
<pre><code>load(url("http://www.produnis.de/R/NEXT-Q0-OR.RData"))
require(rms)
mydata <- NEXT.or
fit <- lrm(QuitJob~QD,data=mydata);fit
</code></pre>
<p>I would like to predict the rank in "QuitJob" (1=won't quit ... | 70,715 |
<p>There have some extensions to Dirichlet process. One is Hierarchical Dirichlet process, and another is Nested Dirichlet Process. What are the differences between these two?</p>
<p>I once read the paper of <a href="http://ba.stat.cmu.edu/journal/2013/vol08/issue02/mueller.pdf" rel="nofollow">Bayesian Nonparametric I... | 70,716 |
<p>I have a multifactorial design (3 treatments x 6 time points, n = 6). One sample per subject. </p>
<p>Because my sample size is small, I've used the KS test for normality - all groups pass. Still, I'm not confident I can assume normality due to small n. </p>
<p>That said, my main issue (I believe) is lack of equal... | 70,717 |
<p>I have a set of around 160 samples from different sediment bodies, or stratigraphic units, which I am attempting to correlate using geochemical analysis of 25 elements for each sample, and a similarity coefficient often applied in correlating volcanic ashes. The literature gives no mention of assumptions of normalit... | 31,800 |
<p>I am interested in doing a sample size calculation for exacerbation rates. I am drawing my data from a negative binomial distribution and my test statistic is based on a negative binomial glm. Currently, I have a code for a power calculation for a given sample size. How do I modify this code to calculate sample size... | 70,718 |
<p>I am looking for some advice as to how I might handle having an unfavorable Levene's test outcome, that is to say a highly significant value. </p>
<p><strong>DOE</strong></p>
<p>Various containers with water were microwaved for 30 seconds under various conditions, the response variable the max temperature. The exp... | 31,803 |
<p>I apologize for the basic question. If $\{p_\theta(x): \theta\in K\subseteq\mathbb{R}\}$ is a smooth family of distributions, then the MLE $\hat{\theta}_n,$ under suitable regularity conditions satisfies $\sqrt{n}(\hat{\theta}_n-\theta)\to\mathcal{N}(0,I(\theta)^{-1})$ where $I(\theta)$ is the Fisher information. We... | 70,719 |
<p>So I'm trying to simulate one winner for each of the 4 NFL divisions, <strong>assuming that all teams are of equal ability</strong>. It's clear that the probability of each team being the division winner is $1/4$.</p>
<p>I wonder if the following simulation strategy works. Instead of simulating each division, I ran... | 31,806 |
<p>I don't want to use SAS-Miner, because I want to create the code to do the optimal binning, even if it's only using entrophy criteria. Is there a code to do so?</p>
<p>Thanks.</p> | 70,720 |
<p>Let's say I have 10 positives out of 1000 observations. I'd like to run <code>glm</code> on the 10 positives and a sample of 10 non-positives (so a total of 20 records in the dataset going into the analysis). To maintain ratios, the weighing scheme is set to 1 for positives and 99 for non-positives, and <code>weight... | 14,146 |
<p>Are there any methods to update the clusters formed in k-way NCut when new data points arrive or any change in the similarity matrix??</p>
<p>thanks</p> | 70,721 |
<p>I have basic question regarding PCA. I am given a task of reducing independent variables in a dataset of about 250,000 enteries and 400 independent variables because some of them are highly correlated. So I want to retain only the most useful ones. I want to perform PCA analysis on it and I have read up but I still ... | 49,282 |
<p>I need to estimate the following conditional variance model</p>
<p>$$
y_t = \sigma_t\,\varepsilon_t
$$</p>
<p>where $y_t$ are the observed data, $\varepsilon_t$ are iid Normal shocks (zero-mean and unit-variance) and where the variance $\sigma_t$ is given by</p>
<p>$$
\sigma_t^2 = \omega+\sum_{j=1}^m\zeta_j\,v_{j... | 3,448 |
<p>I understand the difference between <a href="http://en.wikipedia.org/wiki/Partial_correlation#Semipartial_correlation_.28part_correlation.29" rel="nofollow">partial correlation and semi-partial</a> and zero-order correlation. This terminology makes sense to me in that the partial correlation of X and Y partials out ... | 31,812 |
<p>I recently came a cross Matrix Factorization for classification or recommendations. I was very shocked that it was one of the most effective techniques in the Netflix competition. But actually I think very few people have heard about it. Other techniques like SVM, ANN ... etc are very popular. I was wondering if the... | 31,380 |
<p>I am not really sure about how it behaves when using batch gradient descent in logistic regression.</p>
<p>As we do each iteration, $L(W)$ is getting bigger and bigger, it will jump across the largest point and $L(W)$ is going down. How do I know it without computing $L(W)$ but only knowing old $w$ vector and updat... | 70,722 |
<p>I try to prune one regression tree build with the rpart function in R.
To decide where to prune the tree I used the plotcp function. But I noticed that if I use the same predictor variables and in the same order the plotcp graph always change? How is this possible?
Thank you for your explanation in advance.</p> | 31,816 |
<p>Suppose I have 2 sets:</p>
<p><strong>Set A</strong>: number of items $n= 10$, $\mu = 2.4$ , $\sigma = 0.8$</p>
<p><strong>Set B</strong>: number of items $n= 5$, $\mu = 2$, $\sigma = 1.2$</p>
<p>I can find the combined mean ($\mu$) easily, but how am I supposed to find the combined standard deviation?</p> | 70,723 |
<p>I am struggling to fit alternative count models into my data. I guess my problem is just too many zeros.</p>
<p>This is my data</p>
<pre><code>> summary(smpl)
response predict1 predict2
Min. :0.000 Min. : 1.00 Min. : 22005
1st Qu.:0.000 1st Qu.: 3.00 1st Qu.:... | 88 |
<p>The mixture distribution should be composed of noncentral scaled student t components.</p>
<p>The mean of the mixture distribution can be calculated easily just by weighting the component means according to the mixing law. I also could derieve a tractable analytic formula for the mixture variance by means of the "l... | 930 |
<p>I'm reading about deep learning and that in principles it's a features compression technique and that is why it works. Now my question is why compressing features from 200 or so into 4 is better? How and why does it influence classifiers performance or the construction of the decision boundary?</p>
<p>Look at the f... | 31,818 |
<p>It is known that if $X_i$ are iid, $E(|X_i|) < \infty$ and $E(X_i) = 0$ then $S_n = \sum_1^n X_i$ is a martingale. Suppose all $X_i$ are defined wrt sample space $\Omega$.</p>
<p>I don't understand why $S_n$ is $\sigma(X_1, .. X_n)$-measurable?
Specifically, I am confused about the sample space for $S_n$. To be ... | 70,724 |
<p>I clustered a data set using PAM with a euclidean distance metric and a pearson correlation distance metric. The average silhouette value of the correlation clusters is higher at most points than the euclidean one, but their maximas are at the same number of clusters, and peaks occur at similar cluster numbers.</p>
... | 70,725 |
<p>This is something I was told. </p>
<p>However, it seems to me that parameters of a population can be functions of time, in certain situations, in which case the values of parameters <em>could</em> vary from sample to sample, if the parameter varies with time, unless the samples are all taken at the same time.</p>
... | 70,726 |
<p>I saw this slide with no explanation: <a href="http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CC4QFjAA&url=http%3A%2F%2Fwww.ist.temple.edu%2F~vucetic%2Fcis526fall2004%2Flecture3.ppt&ei=yrSvUp6rG6PIsAS61oLgAw&usg=AFQjCNFVA4ElIWuMdOCF6pqzL_ACl8bEFg&sig2=FCWf... | 49,857 |
<p>There are n data points in R^p space, where p>n. We perform binary classification using linear SVM. As the VC dimension of linear SVM is p+1, it is capable to shatter n points with any label arrangement. Does that mean we should get a 0 training error? However, I rarely get such training error from my experience wor... | 70,727 |
<p>Are there any introductory references or resources that discuss this connection if there is one? I have not been able to find any ones suitable for beginners.</p> | 70,728 |
<p>I found that the default confidence intervals provided by cor.test (Fisher's Z transform according to the help) are quite different from the nonparametric boostrap confidence intervals for the below data. Is it normal this large discrepancy? I am doing something wrong? </p>
<pre><code>dat<-data.frame(x=1:5,y=c(2... | 31,825 |
<p>I'm pretty inexperienced in statistics and am wondering the best way to tackle a couple of simple questions.</p>
<p>I am given the following data on 1,000 hypothetical slot machines:</p>
<p>Attempts made, wins, payout, payout per attempt (win rate * payout)</p>
<p>Some slot machines have over 15,000 attempts, whi... | 31,826 |
<p>In few papers I read that - kernel used doesn't really matter for kernel density estimation but bandwidth of the kernel is the most important factor. But I did not see any mathematical explanation to this and can't completely believe in this. </p> | 70,729 |
<p>Minimum-k (ie, finding the K smallest elements of a dataset) seems like it should be a distributive measure. If I know the smallest k items for two separate datasets, I can compute the smallest k items for the merged dataset by just merging the two sub-results.</p>
<p>However, some lecture slides categorized minimu... | 70,730 |
<p>Given the result $V_{m\times n} \approx W_{m\times k} \cdot H_{k\times n}$, where columns of $V$ are data points and $m$ is data dimension, what is the function by which you assign the data points to the $k$ classes?</p>
<p>Is the usual method to use the normalized columns of $W$ as centroids?</p>
<p>It seems lik... | 70,731 |
<p>I have a question. If I found the effects of a variable is insignificant (p>.05), but the effect size d is 0.2 (small effect).</p>
<p>So I used GPower to calculate the power and I found it is 0.14, which means I need a larger sample size to get a significant results.</p>
<p>So is the hypothesis rejected or support... | 70,732 |
<p>I am reading <a href="http://www.gallup.com/poll/169322/people-worldwide-reporting-lot-positive-emotions.aspx" rel="nofollow">this</a> Gallup poll, and in it there is a "Positive Experience Index" that is used without really explaining how it was created, and therefore what, precisely, it means. From reading the fi... | 31,831 |
<p>I am working on microaggregation problem, where all records are clustered in groups of a minimum size. We must optimize the sum of the squared error (SSE). As I run my algorithm on forest and vlcensus (very large census) datasets, the results for vlcensus are promising, while for the forest they are not satisfying.... | 70,733 |
<p>I don't have a very solid foundation in measure theory, and this always seems a bit confusing to me so I would appreciate any help.</p>
<p>We are given
$
E \left( X_{n+1} | Y_0,\ldots,Y_n \right) = X_n.
$
Prove or disprove
$
E \left( X_{n+1} | Y_0^2,\ldots,Y_n^2 \right) = X_n
$</p>
<p>I am thinking that if $F=\si... | 12,146 |
<p>Hopefully this is an easily answerable question, though I've been searching for an answer without any luck.</p>
<p>I've run a 2x2x2x2 mixed-design ANOVA with two within-subjects and two between-subjects factors. I'm trying to follow up the only significant interaction, which is between the two between-subjects fact... | 70,734 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.