Skip to main content

Postprocessing of MCMC

South, Leah F., Riabiz, Marina, Teymur, Onur, Oates, Chris J. (2022) Postprocessing of MCMC. Annual Review of Statistics and Its Application, 9 (1). ISSN 2326-831X. E-ISSN 2326-831X. (doi:10.1146/annurev-statistics-040220-091727) (The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided) (KAR id:90449)

The full text of this publication is not currently available from this repository. You may be able to access a copy if URLs are provided. (Contact us about this Publication)
Official URL
https://doi.org/10.1146/annurev-statistics-040220-...

Abstract

Markov chain Monte Carlo is the engine of modern Bayesian statistics, being used to approximate the posterior and derived quantities of interest. Despite this, the issue of how the output from a Markov chain is post-processed and reported is often overlooked. Convergence diagnostics can be used to control bias via burn-in removal, but these do not account for (common) situations where a limited computational budget engenders a bias-variance trade-off. The aim of this article is to review state-of-the-art techniques for post-processing Markov chain output. Our review covers methods based on discrepancy minimisation, which

directly address the bias-variance trade-off, as well as general-purpose control variate methods for approximating expected quantities of interest.

Item Type: Article
DOI/Identification number: 10.1146/annurev-statistics-040220-091727
Uncontrolled keywords: bias removal; control variates; Markov chain; Monte Carlo; Stein discrepancy; thinning; variance reduction
Subjects: Q Science > QA Mathematics (inc Computing science) > QA276 Mathematical statistics
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Mathematics, Statistics and Actuarial Science
Depositing User: Onur Teymur
Date Deposited: 29 Sep 2021 08:36 UTC
Last Modified: 21 Feb 2022 16:44 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/90449 (The current URI for this page, for reference purposes)
  • Depositors only (login required):