2
votes
1answer
111 views

Anomaly Detection - a novice question

As I understood from my experimentation on some multivariate data-sets, anomaly detection (AD) heavily depends on the underlying distribution of data. Like, for ex., you can devise a method to detect ...
3
votes
1answer
25 views

Non-informative prior for regression model

I'm looking at p. 355 of Gelman's Bayesian Data Analysis (3rd ed.), for which there is no errata, and I see this: In the normal regression model, a convenient non-informative prior distribution ...
0
votes
0answers
28 views

Transforming time series of different time horizon to stationary

I have a list of monthly time series data with different time periods and different order of integration. I want to transform them all to stationary and a same time period. I noticed that the order ...
0
votes
0answers
3 views

Latent semantic indexing: how to get the low-rank document vectors?

I'm reading the book "Introduction to Information Retrieval" from p.414 to p.416 and I can't understand how the authors computed $C_2$ on p. 415 : http://nlp.stanford.edu/IR-book/pdf/18lsi.pdf They ...
0
votes
0answers
4 views

Understanding Differential Expression Analysis in Microarray Experiments

Can someone please provide me a simple explanation of how differential gene expression analysis works? I know that this method is described in [1] and used in the R package limma. [1] Smyth, Gordon ...
0
votes
1answer
24 views

Training and test data for Holt Winters Method

I am new to forecasting. I have weekly data. I want to forecast using the Holt-Winters method. Should I make training and test subsets of the data? How important are the training and test data?
2
votes
2answers
39 views

Forecasting: Different Model for 1 month, 2 month, 6 month forecasts?

I'm still trying to expand my statistics and forecasting technique knowledge. Right now I'm forecasting seasonal contact patterns, so the simplest model I can understand with seasonality is a ...
1
vote
1answer
32 views

Multiplicative errors for linear model

I am trying to figure out the 'standard' way of handling multiplicative error in a linear model, i.e. my model reads: $$ Y_i = (ax_i + b)\varepsilon_i , \quad \varepsilon_i\sim\mathcal{N}(1, ...
0
votes
0answers
11 views

If $X, Y$ are jointly standard normal with correlation $r$, and $a, b$ are constants, what is $p(Y < b | X < a)$?

The application here is interpreting the correlation coefficient $r$ in terms of $X$'s ability to predict $Y$ for extreme values of $X$. For example, if $r = .8$, then what is $p(Y < 0 | X < ...
0
votes
1answer
29 views

Need clarity on alpha, beta, gamma optimization in Triple Exponential Smoothing Forecast

I asked a variation of this question, but I want to be more direct. Take the exact same Triple Exponential Smoothing Model (Holt-Winters with a moving level, trend, and seasonal component)--- Would ...
0
votes
1answer
12 views

Unimodal or bimodal data (MATLAB)?

I am trying to figure out what I did wrong or what I could do to get accurate results. I have n vectors of data, and I am trying to decide whether each dataset is unimodal or bimodal. I assumed that ...
0
votes
0answers
11 views

What is the distribution of inverse Wishart times a Fisher-Von Mises?

Suppose $p$-dimensional vector $x$ takes a Fisher-Von Mises distribution with parameters $\kappa, \mu$, and $M$ is inverse-Wishart with $n$ degrees of freedom and matrix parameter $I_p$, the $p\times ...
0
votes
1answer
77 views

Extracting nonorthogonal sources in ICA/PCA/blind source seperation problem

My problem is essentially a 'blind source separation' problem. I have 3 non-orthogonal sources (or basis functions) and N random linear combinations (mixes) of said sources. My problem is to obtain ...
3
votes
2answers
29 views

Lagged Dependents

I am in a scenario where I am trying to forecast 2014 call volume in a call center based on prior call volumes in 2013 and 2012. How do I difference 2014 call volume, and how do I lag 2012 and 2013 ...
0
votes
1answer
24 views

Avoid negative results in Holt Winters forecasting

I understood that Holt Winters forecasting may results in negative values due to trending. I did reduce trending component value, but still forecast values are negative territory. Our data set will ...
1
vote
1answer
67 views
+50

Generalized Least Squares vs Ordinary Least Squares under a special case

This question regards the problem of Generalized Least Squares. Vectors and matrices will be denoted in bold. Premises. Let $N,K$ be given integers, with $K \gg N > 1$. The transpose of matrix ...
0
votes
0answers
8 views

Heteroskedasticity determined. Standard errors fall with robust?

I have tested my model for heteroskedasticity both graphically and and formally (White and BGP). However, I find correcting for heteroskedasticity my errors actually fall. I recall learning that ...
1
vote
1answer
11 views

Can a single instrument reduce OVB and measurement error in your model?

My endogenous variable is very likely to be measured with error, but also confounded by unobservables. I have an instrument that rises with the value of the endogenous variable, at the same time it is ...
0
votes
0answers
4 views

Combinef in R HTS package- constrain to keep forecasts positive?

When using the combinef function from Rob Hyndman's very useful hts package for forecasting hierarchical and grouped time series, there does not seem to be a way to constrain the optimally combined ...
1
vote
1answer
127 views

Analysis method for count data with unequal sample sizes and categorical predictors?

I have a problem I've been going over and over for months to find the right statistical analysis method. I'm planning to execute the analysis in R, so any mention of appropriate packages is also ...
2
votes
1answer
31 views

Plain english explanation of the Rayleigh distribution?

I need to understand the Rayleigh distribution for a homework assignment in computer networks. Unfortunately, I lack the background knowledge in the field of statistics and probability theory to ...
3
votes
1answer
42 views

Bayes decision theory: Classification error probability

In Bayesian decision theory: Given $\omega_1$ and $\omega_2$ as two classes for classification, $P\left( \omega_1 \right)$ and $P\left( \omega_2\right)$ their prior probabilities, $x$ the feature ...
0
votes
0answers
10 views

Specifying a glmm for panel data

I'm trying to predict counts based on variables sampled on a monthly basis as well as a few that are not related to time. In several places I've read that the MCMCglmm package in R would be ...
25
votes
5answers
713 views

Examples where method of moments can beat maximum likelihood in small samples?

Maximum likelihood estimators (MLE) are asymptotically efficient; we see the practical upshot in that they often do better than method of moments (MoM) estimates (when they differ), even at small ...
5
votes
1answer
52 views

Required: Method of moments fitting routine for the two-parameter generalized Pareto

I am currently using the evd package which fits a two-parameter GPD by maximum likelihood. Since in small samples the MOM is superior to the ML estimation I'd like ...
-1
votes
0answers
6 views

Dataset replication add a new variable in SAS [migrated]

I wanted to ask, (1) in SAS, how can certain dataset be replicated multiple times and make them into one dataset? (2) with a new variable, which indicates the number of replication. for example ...
2
votes
1answer
54 views

Simulation power analysis

How do you perform a power analysis for linear mixed model using R? I measure area of five groups of plants across time. Each group of plants grow under two treatments. I have 12 replicates for each ...
0
votes
1answer
12 views

What's the best way to calculate survival time using outputs from random survival forest

I have built a random survival forest using R package randomForestSRC. The OOB error rate is around 10%. I was wondering whether anyone had some experience in utilizing the outputs from this model ...
0
votes
3answers
87 views

using random forest for missing data imputation in categorical variables ( in R)

I have following type of associated data. The following example step to generate associated variable. p number of variables and n is number of observations. ...
1
vote
0answers
18 views

Finding optimal hyperplane

I have a set of vectors $\{V_i\}$ in $n$-dimensional space. There is a number corresponded to each vector $\alpha_i = f(V_i)$ ($\alpha_i$ can be negative). I want to find a hyperplane which would ...
0
votes
1answer
22 views
+50

Identifying Instrumental Variables to use in tsls function in sem R package

I want to reproduce the Example from Introduction to Econometrics by G. S. Maddala in R using tsls function from ...
3
votes
1answer
141 views

Given a set of sub-graphs, how to infer the underlying graph?

There are two questions I feel puzzled in the recent period. Firstly, if given a set of sub-graphs, which are sampled from the underlying graph, how can I infer the underlying graph given the set ...
0
votes
0answers
16 views

multinomial logistic regression with alternative specific variables

I am working on a multinomial logistic regression problem which involves features from the dependent variable. For example, if the dependent variable is product, the feature includes product specific ...
2
votes
0answers
21 views

goodness of fit llikelihood ratio test with zero values

I have a vector of observed frequencies that have zero values in some cells and a vector of expected frequencies generated by a model. I would like to do a likelihood ratio test rather than a chi ...
2
votes
1answer
41 views

Testing for the significance of the difference-in-differences of adj. R²

Is there a way to test for the significance of the difference-in-differences in adj. R²s in Stata? Let's say I have four subgroups: pre-treatment, pre-control, post-treatment, post-control and I want ...
0
votes
2answers
185 views

Moralization and triangulation on belief networks

Assume that I have a belief network with a set of nodes. In order to create a valid junction tree I have to moralize the graph. Assume now that I have nodes with more than 2 parents (e.g 3 parents) ...
4
votes
1answer
48 views

Does it make sense to perform a one-tailed Kolmogorov-Smirnov test?

Is it meaningful and possible to perform a one-tailed KS test? What would the null hypothesis of such a test be? Or is the KS test inherently a two-tailed test? I would benefit from an answer that ...
10
votes
1answer
163 views

Estimating variance of center-censored Normal samples

I have normally-distributed processes from which I get small samples (n typically 10-30) that I want to use to estimate variance. But frequently the samples are so close together that we can't ...
3
votes
1answer
81 views

Why doesn't regression results change after bootstrap?

I learned bootstrap is used to treat non-normality of residual and it basically does resampling. I did bootstrapping on Stata and compared the result with normal regression. ...
3
votes
2answers
239 views

Clustering of points based on vector feature similarities in R

I have as an input a number of points that I need to partition into clusters. Each point has a number of features that are ideally to be used to find the similarity between each point and the others. ...
0
votes
0answers
6 views

Combining Standard deviation 3 groups

Is this the correct formula for combining the standard deviations of 3 groups? $$ \sigma = \sqrt{\frac{n_1\sigma_1^2 + n_2 \sigma_2^2+ n_3 \sigma_3^2+ n_1(\mu_1-\mu)^2 ...
2
votes
0answers
18 views

Correlation vs Partial Correlation to explore relationship in data

I have a cognitive architecture solving a set of tasks. Also I have data of human subjects solving the same set of tasks. Now I want to see whether I can find a relationship between the performance ...
1
vote
1answer
198 views

how does Cox proportional hazards model deal with time-dependent variables

When considering time dependent data in survival analysis, you have multiple start-stop times for an individual subject with measurements for the covariates. If each season has a different size (for ...
0
votes
0answers
9 views

Down-sampling with building models (specifically random forests)

I was wondering if anyone had ever used down-sampling to build random forests with data that has unbalanced classes. Basically down-sampling samples (with replacement) x*min from the population where ...
0
votes
1answer
21 views

Multivariate Time Series

I am trying to learn multivariate time series using R. I have two time series and I want to see if I could use one of those to predict the other one, and after that check if the model holds or there ...
0
votes
0answers
9 views

Select range of dates above and below another date [migrated]

Consider these small dataframes a and b: ...
-1
votes
0answers
23 views

T-Statistics - What is it actually infering? [on hold]

Hoping you can help me here. Could you explain the T-Statistics a little more to me? So far I understand how to compute it and why you compute it however I don't understand why it is considered of ...
3
votes
1answer
150 views

How to compare the effect of age on brain volume across different brain regions when the regions are of different sizes?

I am working on analyzing volumetric information extracted from the brain. In particular we would like to compare the strength of effects across three different regions (say x, y, and z). However, the ...
6
votes
3answers
321 views

Do the pdf and the pmf and the cdf contain the same information?

Do the pdf and the pmf and the cdf contain the same information? For me the pdf gives the whole probability to a certain point(basically the area under the probability). The pmf give the probability ...
7
votes
3answers
218 views

How is cross validation different from data snooping?

I just finished "An Introduction to Statistical Learning". I wondered whether using cross-validation to find the best tuning parameters for various machine learning techniques is different from data ...

15 30 50 per page