OpenBUGS

by Andrew Thomas

Published 7 December 2007
BUGS (Bayesian inference Using Gibbs Sampling) is a piece of computer software for the Bayesian analysis of complex statistical models using Markov chain Monte Carlo (MCMC) methods. It has been developing and rapidly growing in popularity for a number of years, and is probably best known in its WinBUGS incarnations. OpenBUGS is the newest innovation: an Open Source version, which can run on Windows and Linux, as well as from inside the increasingly popular R statistical package.This would be the first book on this subject, providing a hands-on guide to this important Open Source software project. The contents will cover all the essential issues of OpenBUGS and the BUGS language as a whole, to enable the reader to perform their own statistical inferences on different complex models. After introducing the area by defining OpenBUGS, the problem domain of complex statistical models is described, and the selection process for finding the most appropriate simulation algorithm for particular models, is explained.
Finally, the reader's experience will be reinforced with real life case studies and an already well established website that will contain the complete source code of the software.