Neyman–Pearson lemma for Bayes factors

DOI10.1080/03610926.2021.2007265MaRDI QIDQ6170110zbMATH OpenOpenAlexFDO

Authors Andrew Fowlie

Publication date 12 July 2023

Published in Communications in Statistics: Theory and Methods (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2110.15625

zbMATH Keywords

\(p\)-valuesBayes factorstesting

Mathematics Subject Classification ID

Statistics (62-XX)

Abstract: We point out that the Neyman-Pearson lemma applies to Bayes factors if we consider expected type-1 and type-2 error rates. That is, the Bayes factor is the test statistic that maximises the expected power for a fixed expected type-1 error rate. For Bayes factors involving a simple null hypothesis, the expected type-1 error rate is just the completely frequentist type-1 error rate. Lastly we remark on connections between the Karlin-Rubin theorem and uniformly most powerful tests, and Bayes factors. This provides frequentist motivations for computing the Bayes factor and could help reconcile Bayesians and frequentists.

Cites work

• Abandon Statistical Significance
• Bayes Factors
• Bayesian (mean) most powerful tests
• Could Fisher, Jeffreys and Neyman have agreed on testing? (With comments and a rejoinder).
• Frequentist probability and frequentist statistics
• scientific article; zbMATH DE number 2166302 (Why is no real title available?)
• scientific article; zbMATH DE number 3276287 (Why is no real title available?)
• scientific article; zbMATH DE number 3195782 (Why is no real title available?)
• scientific article; zbMATH DE number 3080020 (Why is no real title available?)
• IX. On the problem of the most efficient tests of statistical hypotheses
• Likelihood ratio statistics based on an integrated likelihood
• Prior distributions for objective Bayesian analysis
• Rejection odds and rejection ratios: a proposal for statistical practice in testing hypotheses
• Saddle-point Methods for the Multinomial Distribution
• Significance Level and Power
• The Bayes/Non-Bayes Compromise and the Multinomial Distribution
• The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses
• The Selection of Prior Distributions by Formal Rules
• Unified frequentist and Bayesian testing of a precise hypothesis. With comments by Dennis V. Lindley, Thomas A. Louis and David Hinkley and a rejoinder by the authors

This page was built for publication: Neyman–Pearson lemma for Bayes factors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6170110)