Bayesian Cognitive Modeling: A Practical Course

by Michael D. Lee, Eric-Jan Wagenmakers

Bayesian inference has become a standard method of analysis in many fields of science. Students and researchers in experimental psychology and cognitive science, however, have failed to take full advantage of the new and exciting possibilities that the Bayesian approach affords. Ideal for teaching and self study, this book demonstrates how to do Bayesian modeling. Short, to-the-point chapters offer examples, exercises, and computer code (using WinBUGS or JAGS, and supported by Matlab and R), with additional support available online. No advance knowledge of statistics is required and, from the very start, readers are encouraged to apply and adjust Bayesian analyses by themselves. The book contains a series of chapters on parameter estimation and model selection, followed by detailed case studies from cognitive science. After working through this book, readers should be able to build their own Bayesian models, apply the models to their own data, and draw their own conclusions.

Genres: Science, Mathematics

446 pages, Kindle Edition

First published December 31, 2013

Reviews

[Jonas Lindeløv · Rating 5 out of 5]

This book is great as an introduction to BUGS code which is a very powerful way to build all kinds of Bayesian models. It does so by going through many exercises and examples rather than presenting a lot of abstract text. This is a great way to learn and it is excellently executed in this book.

For me, this was just what I needed back when I read it as a relative novice and it showed the endless potential in what you can model. This is it's strength: an easy introduction to (potentially) advanced modelibg. However, if you just want to do Bayesian versions of classical tests such as t-tests, ANOVA, and orher GLMM, it is probably less of interest.