Variational approximations in Bayesian model selection for finite mixture distributions

McGrory, C. A. and Titterington, D. M. (2007) Variational approximations in Bayesian model selection for finite mixture distributions. Computational Statistics & Data Analysis, 51 11: 5352-5367. doi:10.1016/j.csda.2006.07.020


Author McGrory, C. A.
Titterington, D. M.
Title Variational approximations in Bayesian model selection for finite mixture distributions
Journal name Computational Statistics & Data Analysis   Check publisher's open access policy
ISSN 0167-9473
Publication date 2007-01-01
Sub-type Article (original research)
DOI 10.1016/j.csda.2006.07.020
Volume 51
Issue 11
Start page 5352
End page 5367
Total pages 16
Place of publication Amsterdam
Publisher Elsevier Science Bv
Language eng
Abstract Variational methods, which have become popular in the neural computing/machine learning literature, are applied to the Bayesian analysis of mixtures of Gaussian distributions. It is also shown how the deviance information criterion, (DIC), can be extended to these types of model by exploiting the use of variational approximations. The use of variational methods for model selection and the calculation of a DIC are illustrated with real and simulated data. The variational approach allows the simultaneous estimation of the component parameters and the model complexity. It is found that initial selection of a large number of components results in superfluous components being eliminated as the method converges to a solution. This corresponds to an automatic choice of model complexity. The appropriateness of this is reflected in the DIC values. (C) 2006 Elsevier B.V. All rights reserved.
Keyword Computer Science, Interdisciplinary Applications
Statistics & Probability
Bayesian analysis
deviance information criterion (DIC)
mixtures
variational approximations
Deviance Information Criterion
Hidden Markov-models
Graphical Models
Unknown Number
Components
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Unknown

Document type: Journal Article
Sub-type: Article (original research)
Collection: Journal Article Import (ISI/CVs)
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 53 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 68 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Tue, 19 Feb 2008, 00:28:00 EST