Estimating the evidence -- a review
Methodology
2011-11-09 v1
Abstract
The model evidence is a vital quantity in the comparison of statistical models under the Bayesian paradigm. This paper presents a review of commonly used methods. We outline some guidelines and offer some practical advice. The reviewed methods are compared for two examples; non-nested Gaussian linear regression and covariate subset selection in logistic regression.
Cite
@article{arxiv.1111.1957,
title = {Estimating the evidence -- a review},
author = {Nial Friel and Jason Wyse},
journal= {arXiv preprint arXiv:1111.1957},
year = {2011}
}
Comments
21 pages. To appear in a special issue, "All Models Are Wrong...", of Statistica Neerlandica