Translator Disclaimer
February 2007 A General Framework for the Parametrization of Hierarchical Models
Omiros Papaspiliopoulos, Gareth O. Roberts, Martin Sköld
Statist. Sci. 22(1): 59-73 (February 2007). DOI: 10.1214/088342307000000014

Abstract

In this paper, we describe centering and noncentering methodology as complementary techniques for use in parametrization of broad classes of hierarchical models, with a view to the construction of effective MCMC algorithms for exploring posterior distributions from these models. We give a clear qualitative understanding as to when centering and noncentering work well, and introduce theory concerning the convergence time complexity of Gibbs samplers using centered and noncentered parametrizations. We give general recipes for the construction of noncentered parametrizations, including an auxiliary variable technique called the state-space expansion technique. We also describe partially noncentered methods, and demonstrate their use in constructing robust Gibbs sampler algorithms whose convergence properties are not overly sensitive to the data.

Citation

Download Citation

Omiros Papaspiliopoulos. Gareth O. Roberts. Martin Sköld. "A General Framework for the Parametrization of Hierarchical Models." Statist. Sci. 22 (1) 59 - 73, February 2007. https://doi.org/10.1214/088342307000000014

Information

Published: February 2007
First available in Project Euclid: 1 August 2007

zbMATH: 1246.62195
MathSciNet: MR2408661
Digital Object Identifier: 10.1214/088342307000000014

Rights: Copyright © 2007 Institute of Mathematical Statistics

JOURNAL ARTICLE
15 PAGES


SHARE
Vol.22 • No. 1 • February 2007
Back to Top