Default Bayesian model determination methods for generalised linear mixed models

Overstall, Anthony M. and Forster, Jonathan J. (2009) Default Bayesian model determination methods for generalised linear mixed models. Southampton, UK, Southampton Statistical Sciences Research Institute, 25pp. (S3RI Methodology Working Papers, (doi:10.1016/j.csda.2010.03.008) , M09/01).


[img] PDF - Author's Original
Download (1412Kb)


In this paper, we consider a default strategy for fully Bayesian model determination for GLMMs. We address the two key issues of default prior specification and computation. In particular, we extend a concept of unit information to the priors for the parameters of a GLMM. We rely on a combination of MCMC and Laplace approximations to compute approximations to the posterior model probabilities and then further approximate these posterior model probabilities using bridge sampling. We apply our strategy to two examples.

Item Type: Monograph (Working Paper)
Digital Object Identifier (DOI): doi:10.1016/j.csda.2010.03.008
Keywords: unit information priors, bridge sampling, mcmc, laplace approximation
Subjects: H Social Sciences > HA Statistics
Divisions : University Structure - Pre August 2011 > Southampton Statistical Sciences Research Institute
ePrint ID: 64862
Accepted Date and Publication Date:
19 January 2009Submitted
Date Deposited: 20 Jan 2009
Last Modified: 19 Apr 2016 12:04

Actions (login required)

View Item View Item

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics