Open Access
Translator Disclaimer
September 2007 Splitting and merging components of a nonconjugate Dirichlet process mixture model
Sonia Jain, Radford M. Neal
Bayesian Anal. 2(3): 445-472 (September 2007). DOI: 10.1214/07-BA219

Abstract

The inferential problem of associating data to mixture components is difficult when components are nearby or overlapping. We introduce a new split-merge Markov chain Monte Carlo technique that efficiently classifies observations by splitting and merging mixture components of a nonconjugate Dirichlet process mixture model. Our method, which is a Metropolis-Hastings procedure with split-merge proposals, samples clusters of observations simultaneously rather than incrementally assigning observations to mixture components. Split-merge moves are produced by exploiting properties of a restricted Gibbs sampling scan. A simulation study compares the new split-merge technique to a nonconjugate version of Gibbs sampling and an incremental Metropolis-Hastings technique. The results demonstrate the improved performance of the new sampler.

Citation

Download Citation

Sonia Jain. Radford M. Neal. "Splitting and merging components of a nonconjugate Dirichlet process mixture model." Bayesian Anal. 2 (3) 445 - 472, September 2007. https://doi.org/10.1214/07-BA219

Information

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

zbMATH: 1331.62145
MathSciNet: MR2342168
Digital Object Identifier: 10.1214/07-BA219

Rights: Copyright © 2007 International Society for Bayesian Analysis

JOURNAL ARTICLE
28 PAGES


SHARE
Vol.2 • No. 3 • September 2007
Back to Top