Bayesian Methods: A Social and Behavioral Sciences Approach

Front Cover
Taylor & Francis, May 29, 2002 - Mathematics - 459 pages
0 Reviews
Despite increasing interest in Bayesian approaches, especially across the social sciences, it has been virtually impossible to find a text that introduces Bayesian data analysis in a manner accessible to social science students. The Bayesian paradigm is ideally suited to the type of data analysis they will have to perform, but the associated mathematics can be daunting.

Bayesian Methods: A Social and Behavioral Sciences Approach presents the basic principles of Bayesian statistics in a treatment designed specifically for students in the social sciences and related fields. Requiring few prerequisites, it first introduces Bayesian statistics and inference with detailed descriptions of setting up a probability model, specifying prior distributions, calculating a posterior distribution, and describing the results. This is followed by explicit guidance on assessing model quality and model fit using various diagnostic techniques and empirical summaries. Finally, it introduces hierarchical models within the Bayesian context, which leads naturally to Markov Chain Monte Carlo computing techniques and other numerical methods.

The author emphasizes practical computing issues, includes specific details for Bayesian model building and testing, and uses the freely available R and BUGS software for examples and exercise problems. The result is an eminently practical text that is comprehensive, rigorous, and ideally suited to teaching future empirical social scientists.

What people are saying - Write a review

We haven't found any reviews in the usual places.

About the author (2002)

Jeff Gill is an associate professor of Political Science at the University of California, Davis. His current research is focused on projects such as Bayesian hierarchical models, MCMC theory, budgetary behavior in the House of Representatives, Congress, generalized linear model theory, and simulation techniques. He recently completed a book entitled Numerical Issues in Statistical Computing for the Social Scientist (with Altman and McDonald).