Translator Disclaimer
June 2007 Identifying outliers in Bayesian hierarchical models: a simulation-based approach
E. C. Marshall, D. J. Spiegelhalter
Bayesian Anal. 2(2): 409-444 (June 2007). DOI: 10.1214/07-BA218

Abstract

A variety of simulation-based techniques have been proposed for detection of divergent behaviour at each level of a hierarchical model. We investigate a diagnostic test based on measuring the conflict between two independent sources of evidence regarding a parameter: that arising from its predictive prior given the remainder of the data, and that arising from its likelihood. This test gives rise to a $p$-value that exactly matches or closely approximates a cross-validatory predictive comparison, and yet is more widely applicable. Its properties are explored for normal hierarchical models and in an application in which divergent surgical mortality was suspected. Since full cross-validation is so computationally demanding, we examine full-data approximations which are shown to have only moderate conservatism in normal models. A second example concerns criticism of a complex growth curve model at both observation and parameter levels, and illustrates the issue of dealing with multiple $p$-values within a Bayesian framework. We conclude with the proposal of an overall strategy to detecting divergent behaviour in hierarchical models.

Citation

Download Citation

E. C. Marshall. D. J. Spiegelhalter. "Identifying outliers in Bayesian hierarchical models: a simulation-based approach." Bayesian Anal. 2 (2) 409 - 444, June 2007. https://doi.org/10.1214/07-BA218

Information

Published: June 2007
First available in Project Euclid: 22 June 2012

zbMATH: 1331.62032
MathSciNet: MR2312289
Digital Object Identifier: 10.1214/07-BA218

Rights: Copyright © 2007 International Society for Bayesian Analysis

JOURNAL ARTICLE
36 PAGES


SHARE
Vol.2 • No. 2 • June 2007
Back to Top