Is the p-value a good measure of evidence? Asymptotic consistency criteria

DOI10.1016/J.SPL.2012.02.018zbMATH Open1239.62006arXiv1111.4821OpenAlexW2075117686MaRDI QIDQ426698FDOQ426698

Authors: M. Grendár 

Publication date: 11 June 2012

Published in: Statistics \& Probability Letters (Search for Journal in Brave)

Abstract: What are the criteria that a measure of statistical evidence should satisfy? It is argued that a measure of evidence should be consistent. Consistency is an asymptotic criterion: the probability that if a measure of evidence in data strongly testifies against a hypothesis H, then H is indeed not true, should go to one, as more and more data appear. The p-value is not consistent, while the ratio of likelihoods is.

Full work available at URL: https://arxiv.org/abs/1111.4821

Recommendations

• scientific article; zbMATH DE number 850149
• Significance testing, \(p\)-values and the principle of total evidence
• Two-Tailed p-Values and Coherent Measures of Evidence
• How the Maximal Evidence of P-Values Against Point Null Hypotheses Depends on Sample Size
• P-values based on approximate conditioning and \(p^{*}\)
• Theoretical and empirical distributions of the \(p\) value

zbMATH Keywords

statistical evidenceratio of likelihoodsunconditional consistency

Mathematics Subject Classification ID

Foundations and philosophical topics in statistics (62A01)

Cites Work

• Title not available (Why is that?)
• Statistical Evidence
• Bayes Factors
• Calibration of \(\rho \) values for testing precise null hypotheses
• Title not available (Why is that?)
• Simultaneous Test Procedures--Some Theory of Multiple Comparisons
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the Probability of Observing Misleading Statistical Evidence
• Title not available (Why is that?)
• On defining \(P\)-values

Cited In (6)

• Pseudo-likelihood, explanatory power, and Bayes's theorem [Comment on: ``A likelihood paradigm for clinical trials]
• A classical measure of evidence for general null hypotheses
• Consistency of \(p\)-values obtained by averaging over nuisance parameters
• Inference after checking multiple Bayesian models for data conflict and applications to mitigating the influence of rejected priors
• Critique of p‐Values
• Maximum probability/entropy translating of contiguous categorical observations into frequencies

This page was built for publication: Is the \(p\)-value a good measure of evidence? Asymptotic consistency criteria

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q426698)