Nothing Special   »   [go: up one dir, main page]

Open Access
December 2006 Deviance information criteria for missing data models
G. Celeux, F. Forbes, C. P. Robert, D. M. Titterington
Bayesian Anal. 1(4): 651-673 (December 2006). DOI: 10.1214/06-BA122

Abstract

The deviance information criterion (DIC) introduced by Spiegelhalter et al.(2002) for model assessment and model comparison is directly inspired by linear and generalised linear models, but it is open to different possible variations in the setting of missing data models, depending in particular on whether or not the missing variables are treated as parameters. In this paper, we reassess the criterion for such models and compare different DIC constructions, testing the behaviour of these various extensions in the cases of mixtures of distributions and random effect models.

Citation Download Citation

G. Celeux. F. Forbes. C. P. Robert. D. M. Titterington. "Deviance information criteria for missing data models." Bayesian Anal. 1 (4) 651 - 673, December 2006. https://doi.org/10.1214/06-BA122

Information

Published: December 2006
First available in Project Euclid: 22 June 2012

zbMATH: 1331.62329
MathSciNet: MR2282197
Digital Object Identifier: 10.1214/06-BA122

Keywords: completion , deviance , DIC , EM algorithm , MAP , mixture model , model comparison , random effect model

Rights: Copyright © 2006 International Society for Bayesian Analysis

Vol.1 • No. 4 • December 2006
Back to Top