Open Access
Translator Disclaimer
February 2013 Adaptive Gibbs samplers and related MCMC methods
Krzysztof Łatuszyński, Gareth O. Roberts, Jeffrey S. Rosenthal
Ann. Appl. Probab. 23(1): 66-98 (February 2013). DOI: 10.1214/11-AAP806

Abstract

We consider various versions of adaptive Gibbs and Metropolis-within-Gibbs samplers, which update their selection probabilities (and perhaps also their proposal distributions) on the fly during a run by learning as they go in an attempt to optimize the algorithm. We present a cautionary example of how even a simple-seeming adaptive Gibbs sampler may fail to converge. We then present various positive results guaranteeing convergence of adaptive Gibbs samplers under certain conditions.

Citation

Download Citation

Krzysztof Łatuszyński. Gareth O. Roberts. Jeffrey S. Rosenthal. "Adaptive Gibbs samplers and related MCMC methods." Ann. Appl. Probab. 23 (1) 66 - 98, February 2013. https://doi.org/10.1214/11-AAP806

Information

Published: February 2013
First available in Project Euclid: 25 January 2013

zbMATH: 1263.60067
MathSciNet: MR3059204
Digital Object Identifier: 10.1214/11-AAP806

Subjects:
Primary: 60J05 , 65C05
Secondary: 62F15

Keywords: adaptive MCMC , Gibbs sampling , MCMC estimation

Rights: Copyright © 2013 Institute of Mathematical Statistics

JOURNAL ARTICLE
33 PAGES


SHARE
Vol.23 • No. 1 • February 2013
Back to Top