Category: Blog

Linear models for quantiles

Published July 24, 2015 by Corson N. Areshenkoff

Most statistical models used in experimental psychology are designed to estimate the mean of a response variable given some set of predictors. This is all well and good when errors are largely symmetric and our predictors are expected to primarily … Read more

Comments closed

Factorizing motion trajectories

Published May 1, 2015 by Corson N. Areshenkoff

It’s difficult to analyze trajectory data, since the “object” of analysis (the trajectory) can’t easily be summarized by a single value without throwing away large amounts of information, unless you already know what you’re looking for (e.g. if you’re looking … Read more

Comments closed

Publication quality plots in R

Published February 16, 2015 by Corson N. Areshenkoff

The trend is slowly moving towards interactive plotting and reproducible research with e.g. knitR (which I fully support), but it seems to me that R/ggplot2 is still the best plotting system for static graphics. That said, ggplot’s stock settings … Read more

Comments closed

Mixture modelling for reaction time distributions

Published January 1, 2015 by Corson N. Areshenkoff

I’m swimming in reaction time data at the moment. My usual approach to analyzing reaction times in cognitive psychology is through some sort of normal random effects model of the log or inverse RT. This is really only good for … Read more

Comments closed

A note on the CLT

Published December 2, 2014 by Corson N. Areshenkoff

Introductory statistics courses (particularly when they’re taught outside of the statistics department) often gloss over the details of the central limit theorem, describing it only as something that let’s you do t-tests without worrying about normality. I recently came across … Read more

Comments closed

Echo-state networks in R

Published December 1, 2014 by Corson N. Areshenkoff

Echo-state networks (ESN’s) are a class of neural network designed to work with temporal data. I normally don’t like neural networks, since they rarely actually tell you anything interpretable about your data (and they pretty much just do regression anyway), … Read more

Comments closed

Robust t-regression

Published November 1, 2014 by Corson N. Areshenkoff

I’ve been experimenting with techniques for robust regression, and I thought that it would be a fun excercise to implement a robust variant of the simple linear regression model based on the t-distribution.

Motivation

The term “outlier” is used very … Read more

Comments closed

Doing meta-analysis II

Published October 2, 2014 by Corson N. Areshenkoff

I recently collaborated on a meta-analysis investigating the effects of blast-related (i.e. *BOOM*) mild traumatic brain injury (mTBI) on cognitive performance (Karr, et al. 2014). Each of the eight included studies used control and mTBI groups, and reported means and … Read more

Comments closed

Doing meta-analysis I

Published October 1, 2014 by Corson N. Areshenkoff

Meta-analysis is, broadly, a set of statistical models for combining the results of several research studies in order to summarize the literature, or estimate an effect more precisely than can be done by a single study alone. The actual process … Read more

Comments closed