Asymptotic equivalence of empirical likelihood and Bayesian MAP

DOI10.1214/08-AOS645zbMATH Open1173.62014arXiv0908.3397OpenAlexW3098429533MaRDI QIDQ834346FDOQ834346

Authors: N. E. Zubov 

Publication date: 19 August 2009

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: In this paper we are interested in empirical likelihood (EL) as a method of estimation, and we address the following two problems: (1) selecting among various empirical discrepancies in an EL framework and (2) demonstrating that EL has a well-defined probabilistic interpretation that would justify its use in a Bayesian context. Using the large deviations approach, a Bayesian law of large numbers is developed that implies that EL and the Bayesian maximum a posteriori probability (MAP) estimators are consistent under misspecification and that EL can be viewed as an asymptotic form of MAP. Estimators based on other empirical discrepancies are, in general, inconsistent under misspecification.

Full work available at URL: https://arxiv.org/abs/0908.3397

Recommendations

• Bayesian empirical likelihood
• Higher-order properties of Bayesian empirical likelihood
• Bayesian exponentially tilted empirical likelihood
• Methodology and Algorithms of Empirical Likelihood
• Nonparametric Bayes factors based on empirical likelihood ratios

zbMATH Keywords

estimating equationsright censoringKaplan-Meier estimatorBayesian large deviationsBayesian nonparametric consistencyL-divergencemaximum nonparametric likelihoodPólya sampling

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Large deviations (60F10) Bayesian problems; characterization of Bayes procedures (62C10) Strong limit theorems (60F15)

Cites Work

• Extending the scope of empirical likelihood
• Empirical likelihood and general estimating equations
• Empirical likelihood
• Title not available (Why is that?)
• Empirical likelihood ratio confidence intervals for a single functional
• Information Theoretic Approaches to Inference in Moment Condition Models
• Empirical likelihood for estimating equations with missing values
• Misspecification in infinite-dimensional Bayesian statistics
• Title not available (Why is that?)
• Bayesian empirical likelihood
• Nonparametric Bayesian Estimation of Survival Curves from Incomplete Observations
• An inverse of Sanov's theorem
• On the Bernstein-von Mises theorem with infinite-dimensional parameters
• An Information-Theoretic Alternative to Generalized Method of Moments Estimation
• Title not available (Why is that?)
• On the Asymptotic Behavior of Bayes' Estimates in the Discrete Case
• Bayesian nonparametrics
• Title not available (Why is that?)
• New approaches to Bayesian consistency
• On Bayes procedures
• Combined and least squares empirical likelihood
• Estimation and inference in the case of competing sets of estimating equations
• Bayesian exponentially tilted empirical likelihood
• The method of types [information theory]
• Recursive nonlinear estimation. A geometric approach
• Proper likelihoods for Bayesian analysis
• Title not available (Why is that?)
• Information Theory and Statistics: A Tutorial
• Title not available (Why is that?)
• Connections between entropic and linear projections in asset pricing estimation
• Title not available (Why is that?)
• Extending Doob's consistency theorem to nonparametric densities
• Conditioning by rare sources
• Large-Deviations Theory and Empirical Estimator Choice

Cited In (12)

• Bayesian analysis of longitudinal data via empirical likelihood
• Bayesian empirical likelihood for ridge and Lasso regressions
• Bayesian empirical likelihood estimation of quantile structural equation models
• Bayesian elastic net based on empirical likelihood
• Gaussian Processes and Bayesian Moment Estimation
• Bayesian local influence analysis of general estimating equations with nonignorable missing data
• The computational asymptotics of Gaussian variational inference and the Laplace approximation
• Bayesian empirical likelihood
• Bayesian Estimation and Comparison of Moment Condition Models
• Empty set problem of maximum empirical likelihood methods
• The empirical likelihood prior applied to bias reduction of general estimating equations
• Bayesian learning with Wasserstein barycenters

This page was built for publication: Asymptotic equivalence of empirical likelihood and Bayesian MAP

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q834346)