Header

UZH-Logo

Maintenance Infos

Approximate bayesian model selection with the deviance statistic


Held, Leonhard; Sabanés Bové, Daniel; Gravestock, Isaac (2015). Approximate bayesian model selection with the deviance statistic. Statistical science, 30(2):242-257.

Abstract

Bayesian model selection poses two main challenges: the specification of parameter priors for all models, and the computation of the resulting Bayes factors between models. There is now a large literature on automatic and objective parameter priors in the linear model. One important class are g-priors, which were recently extended from linear to generalized linear models (GLMs). We show that the resulting Bayes factors can be approximated by test-based Bayes factors (Johnson [ Scand. J. Stat. 35 (2008) 354–368]) using the deviance statistics of the models. To estimate the hyperparameter g, we propose empirical and fully Bayes approaches and link the former to minimum Bayes factors and shrinkage estimates from the literature. Furthermore, we describe how to approximate the corresponding posterior distribution of the regression coefficients based on the standard GLM output. We illustrate the approach with the development of a clinical prediction model for 30-day survival in the GUSTO-I trial using logistic regression.

Abstract

Bayesian model selection poses two main challenges: the specification of parameter priors for all models, and the computation of the resulting Bayes factors between models. There is now a large literature on automatic and objective parameter priors in the linear model. One important class are g-priors, which were recently extended from linear to generalized linear models (GLMs). We show that the resulting Bayes factors can be approximated by test-based Bayes factors (Johnson [ Scand. J. Stat. 35 (2008) 354–368]) using the deviance statistics of the models. To estimate the hyperparameter g, we propose empirical and fully Bayes approaches and link the former to minimum Bayes factors and shrinkage estimates from the literature. Furthermore, we describe how to approximate the corresponding posterior distribution of the regression coefficients based on the standard GLM output. We illustrate the approach with the development of a clinical prediction model for 30-day survival in the GUSTO-I trial using logistic regression.

Statistics

Citations

Dimensions.ai Metrics
15 citations in Web of Science®
15 citations in Scopus®
11 citations in Microsoft Academic
Google Scholar™

Altmetrics

Downloads

72 downloads since deposited on 29 Dec 2015
29 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > Epidemiology, Biostatistics and Prevention Institute (EBPI)
Dewey Decimal Classification:610 Medicine & health
Scopus Subject Areas:Physical Sciences > Statistics and Probability
Physical Sciences > General Mathematics
Social Sciences & Humanities > Statistics, Probability and Uncertainty
Language:English
Date:2015
Deposited On:29 Dec 2015 09:29
Last Modified:30 Jul 2020 20:00
Publisher:Institute of Mathematical Statistics
ISSN:0883-4237
OA Status:Hybrid
Free access at:Publisher DOI. An embargo period may apply.
Publisher DOI:https://doi.org/10.1214/14-STS510

Download

Hybrid Open Access

Download PDF  'Approximate bayesian model selection with the deviance statistic'.
Preview
Content: Published Version
Filetype: PDF
Size: 735kB
View at publisher
Download PDF  'Approximate bayesian model selection with the deviance statistic'.
Preview
Content: Published Version
Filetype: PDF
Size: 640kB