The Jeffreys-Lindley paradox and discovery criteria in high energy physics

DOI10.1007/S11229-014-0525-ZzbMATH Open1385.81049arXiv1310.3791OpenAlexW3106213137MaRDI QIDQ1708765FDOQ1708765

Authors: Robert D. Cousins 

Publication date: 27 March 2018

Published in: Synthese (Search for Journal in Brave)

Abstract: The Jeffreys-Lindley paradox displays how the use of a p-value (or number of standard deviations z) in a frequentist hypothesis test can lead to an inference that is radically different from that of a Bayesian hypothesis test in the form advocated by Harold Jeffreys in the 1930s and common today. The setting is the test of a well-specified null hypothesis (such as the Standard Model of elementary particle physics, possibly with "nuisance parameters") versus a composite alternative (such as the Standard Model plus a new force of nature of unknown strength). The p-value, as well as the ratio of the likelihood under the null hypothesis to the maximized likelihood under the alternative, can strongly disfavor the null hypothesis, while the Bayesian posterior probability for the null hypothesis can be arbitrarily large. The academic statistics literature contains many impassioned comments on this paradox, yet there is no consensus either on its relevance to scientific communication or on its correct resolution. The paradox is quite relevant to frontier research in high energy physics. This paper is an attempt to explain the situation to both physicists and statisticians, in the hope that further progress can be made.

Full work available at URL: https://arxiv.org/abs/1310.3791

Recommendations

• Erratum to: ``The Jeffreys-Lindley paradox and discovery criteria in high energy physics
• scientific article; zbMATH DE number 2084836
• Constraining Exotic Interactions
• Lorentz violation and deep inelastic scattering
• Searching for unexpected physics at the LHC
• Anomaly driven signatures of new invisible physics at the Large Hadron Collider

zbMATH Keywords

Bayesian model selectionJeffreys-Lindley paradox\(p\) valueshigh energy physicsLindley's paradox

Mathematics Subject Classification ID

Minimax procedures in statistical decision theory (62C20) Applications of statistics to physics (62P35) Nuclear physics (81V35) Unified quantum theories (81V22) Other elementary particle theory in quantum theory (81V25)

Cites Work

• Title not available (Why is that?)
• Sampling and Bayes' Inference in Scientific Modelling and Robustness
• Testing Statistical Hypotheses
• A STATISTICAL PARADOX
• Testing a Point Null Hypothesis: The Irreconcilability of P Values and Evidence
• Title not available (Why is that?)
• Bayes Factors
• Principles of Statistical Inference
• Title not available (Why is that?)
• Title not available (Why is that?)
• Outline of a Theory of Statistical Estimation Based on the Classical Theory of Probability
• IX. On the problem of the most efficient tests of statistical hypotheses
• Statistical decision theory and Bayesian analysis. 2nd ed
• A comment on D. V. Lindley's statistical paradox
• Science and Statistics
• A Reference Bayesian Test for Nested Hypotheses and its Relationship to the Schwarz Criterion
• Severe Testing as a Basic Concept in a Neyman–Pearson Philosophy of Induction
• Probability Theory
• Criteria for Bayesian model choice with application to variable selection
• Title not available (Why is that?)
• Bayesian Hypothesis Testing: A Reference Approach
• Testing precise hypotheses. With comments and a rejoinder by the authors
• Title not available (Why is that?)
• Harold Jeffreys's \textit{Theory of probability} revisited
• Title not available (Why is that?)
• Lindley's Paradox
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Reconciling Bayesian and Frequentist Evidence in the One-Sided Testing Problem
• Nobel Lecture: Asymptotic freedom: From paradox to paradigm
• The Large Sample Correspondence between Classical Hypothesis Tests and Bayesian Posterior Odds Tests
• Title not available (Why is that?)
• The nature of statistical evidence.
• Comment: ``Harold Jeffreys's \textit{Theory of probability} revisited
• Comment: The importance of Jeffreys's legacy
• Statistical methods in experimental physics
• Is the Tail Area Useful as an Approximate Bayes Factor?
• Interpreting the Stars in Precise Hypothesis Testing
• A note on the confidence properties of reference priors for the calibration model
• Bayesian inference given data `significant at \(\alpha\)': Tests of point hypotheses
• Comment: ``Harold Jeffreys's \textit{Theory of probability} revisited
• Comment: ``Harold Jeffreys's \textit{Theory of probability} revisited

Cited In (9)

• Erratum to: ``The Jeffreys-Lindley paradox and discovery criteria in high energy physics
• Bayesian model selection with fractional Brownian motion
• Title not available (Why is that?)
• Objective Bayesian approach to the Jeffreys-Lindley paradox
• Statistical problems in particle physics, astrophysics and cosmology. Proceedings of the PHYSTAT05 conference, Oxford, UK, September 12--15, 2005
• Connecting simple and precise P‐values to complex and ambiguous realities (includes rejoinder to comments on “Divergence vs. decision P‐values”)
• A sample size-dependent prior strategy for bridging the Bayesian-frequentist gap in point null hypothesis testing
• History and nature of the Jeffreys-Lindley paradox
• Bayesian perspectives on the discovery of the Higgs particle

This page was built for publication: The Jeffreys-Lindley paradox and discovery criteria in high energy physics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1708765)