Local consistency of Markov chain Monte Carlo methods

DOI10.1007/S10463-013-0403-3MaRDI QIDQ2434135zbMATH OpenOpenAlexFDO

Authors Kengo Kamatani

Publication date 17 February 2014

Published in Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1012.0996

zbMATH Keywords

asymptotic normality

Mathematics Subject Classification ID

Statistical distribution theory (62E99) Order statistics; empirical distribution functions (62G30) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40)

Abstract: In this paper, we introduce the notion of efficiency (consistency) and examine some asymptotic properties of Markov chain Monte Carlo methods. We apply these results to the data augmentation (DA) procedure for independent and identically distributed observations. More precisely, we show that if both the sample size and the running time of the DA procedure tend to infinity the empirical distribution of the DA procedure tends to the posterior distribution. This is a local property of the DA procedure, which may be, in some cases, more helpful than the global properties to describe its behavior. The advantages of using the local properties are the simplicity and the generality of the results. The local properties provide useful insight into the problem of how to construct efficient algorithms.

Recommendations

• Local degeneracy of Markov chain Monte Carlo methods
• Local weak consistency of Markov chain Monte Carlo methods with application to mixture model
• Stability of sequential Markov Chain Monte Carlo methods
• Consistency of Markov chain quasi-Monte Carlo on continuous state spaces
• scientific article; zbMATH DE number 1135106
• Markov chain Monte Carlo methods: theory and practice
• Marginal Markov chain Monte Carlo methods
• Convergence properties of pseudo-marginal Markov chain Monte Carlo algorithms
• Markov chain Monte Carlo for exact inference for diffusions

Cites work

• Asymptotic methods in statistical decision theory
• Asymptotic properties of a stochastic EM algorithm for mixtures with censored data
• Asymptotic properties of Monte Carlo strategies for a cumulative link model
• General Irreducible Markov Chains and Non-Negative Operators
• General state space Markov chains and MCMC algorithms
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 2063864 (Why is no real title available?)
• Markov chains and stochastic stability
• Markov chains for exploring posterior distributions. (With discussion)
• On the computational complexity of MCMC-based estimators in large samples
• On the preservation of local asymptotic normality under information loss
• Seeking efficient data augmentation schemes via conditional and marginal augmentation
• The order of degeneracy of Markov chain Monte Carlo method
• The stochastic EM algorithm: Estimation and asymptotic results

Cited in

(8)

• Efficient strategy for the Markov chain Monte Carlo in high-dimension with heavy-tailed target probability distribution
• Random walk Metropolis algorithm in high dimension with non-Gaussian target distributions
• The order of degeneracy of Markov chain Monte Carlo method
• Local weak consistency of Markov chain Monte Carlo methods with application to mixture model
• Global Consensus Monte Carlo
• Local degeneracy of Markov chain Monte Carlo methods
• Dimension-free mixing times of Gibbs samplers for Bayesian hierarchical models
• Local alignment of Markov chains

This page was built for publication: Local consistency of Markov chain Monte Carlo methods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2434135)