Quasi-reliable estimates of effective sample size

DOI10.1093/IMANUM/DRAA077zbMATH Open1481.65028arXiv1705.03831OpenAlexW2612512107MaRDI QIDQ5024892FDOQ5024892

Authors: Youhan Fang, Yudong Cao, Robert D. Skeel 

Publication date: 27 January 2022

Published in: IMA Journal of Numerical Analysis (Search for Journal in Brave)

Abstract: The efficiency of a Markov chain Monte Carlo algorithm might be measured by the cost of generating one independent sample, or equivalently, the total cost divided by the effective sample size, defined in terms of the integrated autocorrelation time. To ensure the reliability of such an estimate, it is suggested that there be an adequate sampling of state space--- to the extent that this can be determined from the available samples. A possible method for doing this is derived and evaluated.

Full work available at URL: https://arxiv.org/abs/1705.03831

Recommendations

• Multivariate output analysis for Markov chain Monte Carlo
• Markov chains for exploring posterior distributions. (With discussion)
• Markov chain Monte Carlo: can we trust the third significant figure?
• Publication:4892503
• Markov chain Monte Carlo methods: theory and practice

zbMATH Keywords

Markov chain Monte CarloBayesian inferencespectral gapconvergence diagnosticsintegrated autocorrelation time

Mathematics Subject Classification ID

Bayesian inference (62F15) Monte Carlo methods (65C05) Numerical analysis or methods applied to Markov chains (65C40) Sampling theory, sample surveys (62D05)

Cited In (6)

• Title not available (Why is that?)
• Title not available (Why is that?)
• Monte Carlo errors with less errors
• Title not available (Why is that?)
• Sample size and the accuracy of a consistent estimator
• Markov Chain Sampling and the Product Estimator

This page was built for publication: Quasi-reliable estimates of effective sample size

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5024892)