Bayesian model choice via mixture distributions with application to epidemics and population process models

O'Neill, Philip D. and Kypraios, Theodore (2014) Bayesian model choice via mixture distributions with application to epidemics and population process models. Working Paper. University of Nottingham. (Unpublished)

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (141kB) | Preview

Abstract

We consider Bayesian model choice for the setting where the observed data are partially observed realisations of a stochastic population process. A new method for computing Bayes factors is described which avoids the need to use reversible jump approaches. The key idea is to perform inference for a hypermodel in which the competing models are components of a mixture distribution. The method itself has fairly general applicability. The methods are illustrated using simple population process models and stochastic epidemics.

Item Type: Monograph (Working Paper)
Schools/Departments: University of Nottingham UK Campus > Faculty of Science > School of Mathematical Sciences
Depositing User: Kypraios, Dr Theodore
Date Deposited: 28 Nov 2014 09:51
Last Modified: 06 Apr 2016 16:00
URI: http://eprints.nottingham.ac.uk/id/eprint/27861

Actions (Archive Staff Only)

Edit View Edit View