Ioannis Ntzoufras

Publications Page

Gibbs Variable Selection Using BUGS

Ioannis Ntzoufras (2002)

Journal of Statistical Software, Volume 7, Issue 7

available from www.jstatsoft.org  

Abstract

        In this paper we discuss and present in detail the implementation of Gibbs variable selection as defined by  Dellaportas et al. (2000, 2002) using the BUGS software (Spiegelhalter et al., 1996).  The specification of the likelihood, prior and pseudo-prior distributions of the parameters as well as the prior term and model probabilities are described in detail.  Guidance is also provided for the calculation of the posterior probabilities within BUGS environment when the number of models is limited.  We illustrate the application of this methodology in a variety of problems including linear regression, log-linear and binomial response models.

Keywords:  Logistic regression;  Linear regression;  MCMC;  Model selection.

[ Download paper. ]

[Download associated Bugs files here]


Back to Research Page
All Contents Copyright.
Last revised: 15-11-2001