On the Birnbaum argument for the strong likelihood principle

DOI10.1214/13-STS457zbMATH Open1332.62025MaRDI QIDQ257686FDOQ257686

Authors Deborah G. Mayo 

Publication date 17 March 2016

Published in Statistical Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1302.7021

zbMATH Keywords

sufficiencyBirnbaumizationlikelihood principle (weak and strong)sampling theoryweak conditionality

Mathematics Subject Classification ID

Foundations and philosophical topics in statistics (62A01) Sampling theory, sample surveys (62D05) Sufficient statistics and fields (62B05)

Abstract: An essential component of inference based on familiar frequentist notions, such as $p$-values, significance and confidence levels, is the relevant sampling distribution. This feature results in violations of a principle known as the strong likelihood principle (SLP), the focus of this paper. In particular, if outcomes $mathbf{x}^{\ast }$ and $mathbf{y}^{\ast }$ from experiments $E_1$ and $E_2$ (both with unknown parameter $heta$) have different probability models $f_1(cdot),f_2(cdot)$, then even though $f_1(mathbf{x}^{\ast };heta)=c{f}_2(mathbf{y}^{\ast };heta)$ for all $heta$, outcomes $mathbf{x}^{\ast }$ and $mathbf{y}^{\ast }$ may have different implications for an inference about $heta$. Although such violations stem from considering outcomes other than the one observed, we argue this does not require us to consider experiments other than the one performed to produce the data. David Cox [Ann. Math. Statist. 29 (1958) 357-372] proposes the Weak Conditionality Principle (WCP) to justify restricting the space of relevant repetitions. The WCP says that once it is known which $E_i$ produced the measurement, the assessment should be in terms of the properties of $E_i$. The surprising upshot of Allan Birnbaum's [J. Amer. Statist. Assoc. 57 (1962) 269-306] argument is that the SLP appears to follow from applying the WCP in the case of mixtures, and so uncontroversial a principle as sufficiency (SP). But this would preclude the use of sampling distributions. The goal of this article is to provide a new clarification and critique of Birnbaum's argument. Although his argument purports that [(WCP and SP) entails SLP], we show how data may violate the SLP while holding both the WCP and SP. Such cases also refute [WCP entails SLP].

Recommendations

• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Rejoinder: ``On the Birnbaum argument for the strong likelihood principle
• scientific article; zbMATH DE number 4086734
• On principles and arguments to likelihood
• scientific article; zbMATH DE number 94861
• On the foundations of likelihood principle

Cites work

• scientific article; zbMATH DE number 46578 (Why is no real title available?)
• scientific article; zbMATH DE number 3522963 (Why is no real title available?)
• scientific article; zbMATH DE number 3558709 (Why is no real title available?)
• scientific article; zbMATH DE number 1746720 (Why is no real title available?)
• scientific article; zbMATH DE number 2166302 (Why is no real title available?)
• An introduction to Bayesian analysis. Theory and methods.
• Frequentist statistics as a theory of inductive inference
• More on Concepts of Statistical Evidence
• On principles and arguments to likelihood
• On the Foundations of Statistical Inference
• Severe Testing as a Basic Concept in a Neyman–Pearson Philosophy of Induction
• Some Problems Connected with Statistical Inference
• Sufficiency and conditionality
• Testing Statistical Hypotheses
• The case for objective Bayesian analysis
• What does the proof of Birnbaum's theorem prove?

Cited in

(13)

• Interpreting \(p\)-values and confidence intervals using well-calibrated null preference priors
• The concept of sufficiency in conditional frequentist inference
• scientific article; zbMATH DE number 4157669 (Why is no real title available?)
• scientific article; zbMATH DE number 846108 (Why is no real title available?)
• False confidence, non-additive beliefs, and valid statistical inference
• Blending Bayesian and Classical Tools to Define Optimal Sample-Size-Dependent Significance Levels
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• On principles and arguments to likelihood
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Discussion of ``On the Birnbaum argument for the strong likelihood principle
• Non-Experimental Data, Hypothesis Testing, and the Likelihood Principle: A Social Science Perspective
• New two-sided confidence intervals for binomial inference derived using Walley's imprecise posterior likelihood as a test statistic

This page was built for publication: On the Birnbaum argument for the strong likelihood principle

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q257686)