Default Bayesian model determination methods for generalised linear mixed models
Default Bayesian model determination methods for generalised linear mixed models
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.
unit information priors, bridge sampling, mcmc, laplace approximation
Southampton Statistical Sciences Research Institute, University of Southampton
Overstall, Anthony M.
b642132b-31bc-4263-b60a-a38c8c91db52
Forster, Jonathan J.
e3c534ad-fa69-42f5-b67b-11617bc84879
Overstall, Anthony M.
b642132b-31bc-4263-b60a-a38c8c91db52
Forster, Jonathan J.
e3c534ad-fa69-42f5-b67b-11617bc84879
Overstall, Anthony M. and Forster, Jonathan J.
(2009)
Default Bayesian model determination methods for generalised linear mixed models
(S3RI Methodology Working Papers, M09/01)
Southampton, UK.
Southampton Statistical Sciences Research Institute, University of Southampton
25pp.
(doi:10.1016/j.csda.2010.03.008).
(Submitted)
Record type:
Monograph
(Working Paper)
Abstract
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.
Text
Overstall120310.pdf
- Author's Original
More information
Submitted date: 19 January 2009
Keywords:
unit information priors, bridge sampling, mcmc, laplace approximation
Organisations:
Southampton Statistical Research Inst.
Identifiers
Local EPrints ID: 64862
URI: http://eprints.soton.ac.uk/id/eprint/64862
PURE UUID: 6319e399-be98-452e-9ec2-edb9a1d40ab4
Catalogue record
Date deposited: 20 Jan 2009
Last modified: 16 Mar 2024 02:45
Export record
Altmetrics
Contributors
Author:
Anthony M. Overstall
Author:
Jonathan J. Forster
Download statistics
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