A frequentist framework of inductive reasoning

DOI10.1007/S13171-012-0020-XzbMATH Open1281.62027arXivmath/0602377OpenAlexW3099154844WikidataQ57826634 ScholiaQ57826634MaRDI QIDQ2392499FDOQ2392499

Authors: David R. Bickel 

Publication date: 1 August 2013

Published in: Sankhyā. Series A (Search for Journal in Brave)

Abstract: Reacting against the limitation of statistics to decision procedures, R. A. Fisher proposed for inductive reasoning the use of the fiducial distribution, a parameter-space distribution of epistemological probability transferred directly from limiting relative frequencies rather than computed according to the Bayes update rule. The proposal is developed as follows using the confidence measure of a scalar parameter of interest. (With the restriction to one-dimensional parameter space, a confidence measure is essentially a fiducial probability distribution free of complications involving ancillary statistics.) A betting game establishes a sense in which confidence measures are the only reliable inferential probability distributions. The equality between the probabilities encoded in a confidence measure and the coverage rates of the corresponding confidence intervals ensures that the measure's rule for assigning confidence levels to hypotheses is uniquely minimax in the game. Although a confidence measure can be computed without any prior distribution, previous knowledge can be incorporated into confidence-based reasoning. To adjust a p-value or confidence interval for prior information, the confidence measure from the observed data can be combined with one or more independent confidence measures representing previous agent opinion. (The former confidence measure may correspond to a posterior distribution with frequentist matching of coverage probabilities.) The representation of subjective knowledge in terms of confidence measures rather than prior probability distributions preserves approximate frequentist validity.

Full work available at URL: https://arxiv.org/abs/math/0602377

Recommendations

• Coherent frequentism: a decision theory based on confidence sets
• Confidence is epistemic probability for empirical science
• Confidence, Likelihood, Probability
• scientific article; zbMATH DE number 1067781
• Frequency-calibrated belief functions: review and new insights

zbMATH Keywords

coherencemachine learningartificial intelligenceconfidence distributionfoundations of statisticsprior elicitationinterpretation of probabilitysubjective probabilityconfidence posteriorexpert systeminductive reasoningbettingpersonal probability

Mathematics Subject Classification ID

Bayesian inference (62F15) Foundations and philosophical topics in statistics (62A01) Bayesian problems; characterization of Bayes procedures (62C10)

Cites Work

• Applied Asymptotics
• Statistical Evidence
• Title not available (Why is that?)
• Title not available (Why is that?)
• R. A. Fisher and the fiducial argument
• Title not available (Why is that?)
• Confidence and Likelihood^*
• Title not available (Why is that?)
• Title not available (Why is that?)
• The problem of regions
• Estimating the null distribution to adjust observed confidence levels for genome-scale screening
• Notions of Limiting P Values Based on Data Depth and Bootstrap
• The formal definition of reference priors
• Combining information from independent sources through confidence distributions
• The Dempster-Shafer calculus for statisticians
• Ancillaries and conditional inference (with comments and rejoinder)
• Some Problems Connected with Statistical Inference
• On the Probability of Observing Misleading Statistical Evidence
• Bayes and likelihood calculations from confidence intervals
• Title not available (Why is that?)
• Statistical Methods for Eliciting Probability Distributions
• Title not available (Why is that?)
• Subjective Bayesian analysis: principles and practice
• Strong matching of frequentist and Bayesian parametric inference
• Title not available (Why is that?)
• Statistical inference in science
• Theory of statistics
• Prior inferences for posterior judgements
• Title not available (Why is that?)
• Decision making under uncertainty using imprecise probabilities
• Information optimality and Bayesian modelling
• The Uncertain Reasoner's Companion
• Bayes' Method for Bookies
• Estimation of accuracy in testing
• What is a statistical model? (With comments and rejoinder).
• A betting interpretation for probabilities and Dempster-Shafer degrees of belief
• Probability matching priors: Higher order asymptotics
• Uncertain Inference
• Conditioning, Likelihood, and Coherence: A Review of Some Foundational Concepts
• Controlling the degree of caution in statistical inference with the Bayesian and frequentist approaches as opposite extremes
• The strength of statistical evidence for composite hypotheses: inference to the best explanation
• Coherent Frequentism: A Decision Theory Based on Confidence Sets
• Statistical Inference: Likelihood to Significance
• Title not available (Why is that?)
• Conditional Confidence Statements and Confidence Estimators
• Empirical Bayes interval estimates that are conditionally equal to unadjusted confidence intervals or to default prior credibility intervals
• R. A. Fisher in the 21st century. Invited paper presented at the 1996 R. A. Fisher lecture. (With comments).
• Linear utility theory for belief functions
• An algebraic theory for statistical information based on the theory of hints
• Title not available (Why is that?)
• The Equivalence of Constrained and Weighted Designs in Multiple Objective Design Problems
• Objective Bayesianism, Bayesian conditionalisation and voluntarism
• On finitely additive priors, coherence, and extended admissibility
• Coherent inference from improper priors and from finitely additive priors
• Title not available (Why is that?)
• Title not available (Why is that?)
• R.A. Fisher: A True Bayesian?
• Confidence, posterior probability, and the Buehler example
• A Note on a Reformulation of the S-Method of Multiple Comparison
• A note on Dempster-Shafer recombination of confidence distributions
• On hacking's fiducial theory of inference
• Avoiding foregone conclusions: Geometric and foundational analysis of paradoxes of finite additivity. (With comments)

Cited In (9)

• A predictive approach to measuring the strength of statistical evidence for single and multiple comparisons
• A prior-free framework of coherent inference and its derivation of simple shrinkage estimators
• Title not available (Why is that?)
• Estimating the null distribution to adjust observed confidence levels for genome-scale screening
• Confidence distributions and empirical Bayes posterior distributions unified as distributions of evidential support
• Maximum entropy derived and generalized under idempotent probability to address Bayes-frequentist uncertainty and model revision uncertainty: an information-theoretic semantics for possibility theory
• Discussion of “Divergence vs. Decision P$$ P $$‐values: A Distinction Worth Making in Theory and Keeping in Practice – or, How Divergence P$$ P $$‐values Measure Evidence Even When Decision P$$ P $$‐values Do Not” by Sander Greenland
• Fisher's disjunction as the principle vindicating \(p\)-values, confidence intervals, and their generalizations: a frequentist semantics for possibility theory
• Confidence intervals, significance values, maximum likelihood estimates, etc. sharpened into Occam’s razors

This page was built for publication: A frequentist framework of inductive reasoning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2392499)