Convergence analysis of the Gibbs sampler for Bayesian general linear mixed models with improper priors

Abstract

Bayesian analysis of data from the general linear mixed model is challenging because any nontrivial prior leads to an intractable posterior density. However, if a conditionally conjugate prior density is adopted, then there is a simple Gibbs sampler that can be employed to explore the posterior density. A popular default among the conditionally conjugate priors is an improper prior that takes a product form with a flat prior on the regression parameter, and so-called power priors on each of the variance components. In this paper, a convergence rate analysis of the corresponding Gibbs sampler is undertaken. The main result is a simple, easily-checked sufficient condition for geometric ergodicity of the Gibbs-Markov chain. This result is close to the best possible result in the sense that the sufficient condition is only slightly stronger than what is required to ensure posterior propriety. The theory developed in this paper is extremely important from a practical standpoint because it guarantees the existence of central limit theorems that allow for the computation of valid asymptotic standard errors for the estimates computed using the Gibbs sampler.

arXiv commentPublished in at http://dx.doi.org/10.1214/12-AOS1052 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (6)

• General state space Markov chains and MCMC algorithms
• A General Framework for the Parametrization of Hierarchical Models
• Markov Chain Monte Carlo: Can We Trust the Third Significant Figure?
• Batch means and spectral variance estimators in Markov chain Monte Carlo
• A spectral analytic comparison of trace-class data augmentation algorithms and their sandwich variants
• Gibbs Sampling, Exponential Families and Orthogonal Polynomials

Cited by in corpus (8)

• Component-Wise Markov Chain Monte Carlo: Uniform and Geometric Ergodicity under Mixing and Composition
• MCMC-Based Inference in the Era of Big Data: A Fundamental Analysis of the Convergence Complexity of High-Dimensional Chains
• Convergence analysis of block Gibbs samplers for Bayesian linear mixed models with $p>N$
• Bayesian Dynamic Quantile Model Averaging
• Convergence Analysis of the Data Augmentation Algorithm for Bayesian Linear Regression with Non-Gaussian Errors
• Convergence analysis of the block Gibbs sampler for Bayesian probit linear mixed models with improper priors
• On the propriety of the posterior of hierarchical linear mixed models with flexible random effects distributions
• Estimating MCMC convergence rates using common random number simulation