Asymptotically minimax Bayesian predictive densities for multinomial models

DOI10.1214/12-EJS700MaRDI QIDQ1950845zbMATH OpenFDO

Authors Fumiyasu Komaki

Publication date 28 May 2013

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1112.0818, https://projecteuclid.org/euclid.ejs/1337951629

zbMATH Keywords

Dirichlet priorreference priorKullback-Leibler divergenceJeffreys priorlatent information prior

Mathematics Subject Classification ID

Bayesian inference (62F15) Bayesian problems; characterization of Bayes procedures (62C10) Minimax procedures in statistical decision theory (62C20)

Abstract: One-step ahead prediction for the multinomial model is considered. The performance of a predictive density is evaluated by the average Kullback-Leibler divergence from the true density to the predictive density. Asymptotic approximations of risk functions of Bayesian predictive densities based on Dirichlet priors are obtained. It is shown that a Bayesian predictive density based on a specific Dirichlet prior is asymptotically minimax. The asymptotically minimax prior is different from known objective priors such as the Jeffreys prior or the uniform prior.

Recommendations

• Asymptotically minimax Bayes predictive densities
• scientific article; zbMATH DE number 2042823
• Improved minimax predictive densities under Kullback-Leibler loss
• Bayesian predictive densities based on latent information priors
• Asymptotically minimax prediction in infinite sequence models

Cites work

• scientific article; zbMATH DE number 3667770 (Why is no real title available?)
• scientific article; zbMATH DE number 3454671 (Why is no real title available?)
• Asymptotic minimax regret for data compression, gambling, and prediction
• Bayesian predictive densities based on latent information priors
• Estimating a Product of Means: Bayesian Analysis with Reference Priors
• Game theory, maximum entropy, minimum discrepancy and robust Bayesian decision theory
• Goodness of prediction fit
• Information optimality and Bayesian modelling
• Jeffreys' prior is asymptotically least favorable under entropy risk
• Minimax redundancy for the class of memoryless sources
• On asymptotic properties of predictive distributions
• Simultaneous prediction of independent Poisson observables

Cited in

(11)

• Asymptotic properties of Bayesian predictive densities when the distributions of data and target variables are different
• scientific article; zbMATH DE number 4174134 (Why is no real title available?)
• On minimax optimality of sparse Bayes predictive density estimates
• Asymptotically minimax prediction in infinite sequence models
• Bayesian predictive densities based on latent information priors
• Bayesian shrinkage estimation for stratified count data
• scientific article; zbMATH DE number 2042823 (Why is no real title available?)
• Bayesian point estimation and predictive density estimation for the binomial distribution with a restricted probability parameter
• Minimax mutual prediction of multinomial random variables
• Asymptotically minimax Bayes predictive densities
• Bayesian shrinkage approaches to unbalanced problems of estimation and prediction on the basis of negative multinomial samples

This page was built for publication: Asymptotically minimax Bayesian predictive densities for multinomial models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1950845)