Predictive distributions that mimic frequencies over a restricted subdomain

DOI10.1007/S10203-020-00281-ZzbMATH Open1444.62032arXiv1909.02499OpenAlexW3025883053MaRDI QIDQ777916FDOQ777916

Giuseppe Sanfilippo, Frank Lad

Publication date: 8 July 2020

Published in: Decisions in Economics and Finance (Search for Journal in Brave)

Abstract: A predictive distribution over a sequence of $N+1$ events is said to be "frequency mimicking" whenever the probability for the final event conditioned on the outcome of the first $N$ events equals the relative frequency of successes among them. Infinitely extendible exchangeable distributions that universally inhere this property are known to have several annoying concomitant properties. We motivate frequency mimicking assertions over a limited subdomain in practical problems of finite inference, and we identify their computable coherent implications. We provide some computed examples using reference distributions, and we introduce computational software to generate any specification. The software derives from an inversion of the finite form of the exchangeability representation theorem. Three new theorems delineate the extent of the usefulness of such distributions, and we show why it may not be appropriate to extend the frequency mimicking assertions for a specified value of $N$ to any arbitrary larger size of $N$. The constructive results identify the source and structure of "adherent masses" in the limit of a sequence of finitely additive distributions. Appendices develop a novel geometrical representation of conditional probabilities which illuminate the analysis.

Full work available at URL: https://arxiv.org/abs/1909.02499

Recommendations

• Criticism of the predictive distribution
• Frequentist prediction intervals and predictive distributions
• A class of models for Bayesian predictive inference
• Prediction and asymptotics
• Exchangeability, predictive distributions and parametric models

zbMATH Keywords

probability boundsconditional probabilityprobability elicitation\(A_n\) and \(H_n\) distributionsextendable exchangeability

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Computational methods for problems pertaining to statistics (62-08) Exact distribution theory in statistics (62E15) Markov processes: hypothesis testing (62M02)

Cites Work

• Matlab
• De Finetti's coherence and statistical inference
• Finite forms of de Finetti's theorem on exchangeability
• CHARACTERIZATION OF COHERENT CONDITIONAL PROBABILITIES AS A TOOL FOR THEIR ASSESSMENT AND EXTENSION
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• A generalization of the fundamental theorem of de Finetti for imprecise conditional probability assessments
• Title not available (Why is that?)
• Quasi conjunction, quasi disjunction, t-norms and t-conorms: probabilistic aspects
• Locally strong coherence and inference with lower-upper probabilities
• Title not available (Why is that?)
• Title not available (Why is that?)
• On coherent conditional probabilities and disintegrations
• Basic ideas underlying conglomerability and disintegrability
• Bayesian Nonparametric Survival Analysis
• Coherency conditions for finite exchangeable inference
• Numerical application of the fundamental theorem of prevision
• Title not available (Why is that?)
• Title not available (Why is that?)

Cited In (2)

• Compound conditionals, Fréchet-Hoeffding bounds, and Frank \(t\)-norms
• A special issue on the mathematics of subjective probability

Uses Software

• Matlab
• GAMS

This page was built for publication: Predictive distributions that mimic frequencies over a restricted subdomain

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q777916)