Bayesian Estimation and Comparison of Moment Condition Models

DOI10.1080/01621459.2017.1358172MaRDI QIDQ3121558zbMATH OpenOpenAlexFDO

Authors Siddhartha Chib, M. Shin, Anna Simoni

Publication date 20 March 2019

Published in Journal of the American Statistical Association (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1606.02931

zbMATH Keywords

estimating equationsmarginal likelihoodmisspecificationmodel selection consistencyBernstein-von Mises theoremexponentially tilted empirical likelihood

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Parametric inference under constraints (62F30)

Abstract: In this paper we consider the problem of inference in statistical models characterized by moment restrictions by casting the problem within the Exponentially Tilted Empirical Likelihood (ETEL) framework. Because the ETEL function has a well defined probabilistic interpretation and plays the role of a nonparametric likelihood, a fully Bayesian semiparametric framework can be developed. We establish a number of powerful results surrounding the Bayesian ETEL framework in such models. One major concern driving our work is the possibility of misspecification. To accommodate this possibility, we show how the moment conditions can be reexpressed in terms of additional nuisance parameters and that, even under misspecification, the Bayesian ETEL posterior distribution satisfies a Bernstein-von Mises result. A second key contribution of the paper is the development of a framework based on marginal likelihoods and Bayes factors to compare models defined by different moment conditions. Computation of the marginal likelihoods is by the method of Chib (1995) as extended to Metropolis-Hastings samplers in Chib and Jeliazkov (2001). We establish the model selection consistency of the marginal likelihood and show that the marginal likelihood favors the model with the minimum number of parameters and the maximum number of valid moment restrictions. When the models are misspecified, the marginal likelihood model selection procedure selects the model that is closer to the (unknown) true data generating process in terms of the Kullback-Leibler divergence. The ideas and results in this paper provide a further broadening of the theoretical underpinning and value of the Bayesian ETEL framework with likely far-reaching practical consequences. The discussion is illuminated through several examples.

Recommendations

• Bayesian estimation of state space models using moment conditions
• Bayesian analysis in moment inequality models
• Moment conditions and Bayesian non-parametrics
• Bayesian Model Comparison
• scientific article; zbMATH DE number 5811430
• Bayesian inference and model comparison for random choice structures
• Bayesian comparison of latent variable models: conditional versus marginal likelihoods
• scientific article; zbMATH DE number 849932
• scientific article; zbMATH DE number 1098811
• Bayes estimates in variance components models

Cites work

• scientific article; zbMATH DE number 1220667 (Why is no real title available?)
• A review on empirical likelihood methods for regression
• An Information-Theoretic Alternative to Generalized Method of Moments Estimation
• Asymptotic efficiency in estimation with conditional moment restrictions
• Asymptotic equivalence of empirical likelihood and Bayesian MAP
• Bayesian Pseudo-Empirical-Likelihood Intervals for Complex Surveys
• Bayesian and frequentist confidence intervals arising from empirical-type likelihoods
• Bayesian averaging, prediction and nonnested model selection
• Bayesian empirical likelihood
• Bayesian empirical likelihood for quantile regression
• Bayesian exponentially tilted empirical likelihood
• Bayesian nonparametrics
• Bayesian quantile regression based on the empirical likelihood with spike and slab priors
• Bayesian semiparametric hierarchical empirical likelihood spatial models
• Divergences and duality for estimation and test under moment condition models
• Empirical likelihood
• Empirical likelihood and general estimating equations
• Empirical likelihood based variable selection
• Empirical likelihood for small area estimation
• Empirical likelihood ratio confidence intervals for a single functional
• Empirical likelihood ratio confidence regions
• Empirical-type likelihoods allowing posterior credible sets with frequentist validity: Higher-order asymptotics
• Hamiltonian Monte Carlo sampling in Bayesian empirical likelihood computation
• I-divergence geometry of probability distributions and minimization problems
• Identification problem of the exponential tilting estimator under misspecification
• Large Sample Properties of Generalized Method of Moments Estimators
• Marginal Likelihood From the Metropolis–Hastings Output
• Marginal Likelihood from the Gibbs Output
• Nonparametric Bayes factors based on empirical likelihood ratios
• One-Step Estimators for Over-Identified Generalized Method of Moments Models
• Point estimation with exponentially tilted empirical likelihood
• Semiparametric approach to a random effects quantile regression model
• Tailored randomized block MCMC methods with application to DSGE models
• The Bayesian Choice
• The Bernstein-von Mises theorem under misspecification

Cited in

(29)

• Variational Bayes for Fast and Accurate Empirical Likelihood Inference
• Moment conditions and Bayesian non-parametrics
• scientific article; zbMATH DE number 1396169 (Why is no real title available?)
• Asymptotically valid Bayesian inference in the presence of distributional misspecification in VAR models
• Comparison and robustification of Bayes and Black-Litterman models
• Constrained estimation using penalization and MCMC
• On oracle property and asymptotic validity of Bayesian generalized method of moments
• The Conditional Moment Closure Model
• Limited information likelihood and Bayesian analysis
• Large sample justifications for the Bayesian empirical likelihood
• Variable selection for categorical response: a comparative study
• Semiparametric Bayesian doubly robust causal estimation
• Gaussian Processes and Bayesian Moment Estimation
• Bayesian analysis in moment inequality models
• The empirical saddlepoint estimator
• Which Factors are Risk Factors in Asset Pricing? A Model Scan Framework
• Quasi-Bayesian estimation and inference with control functions
• Bayesian mode and maximum estimation and accelerated rates of contraction
• Bayesian estimation of state space models using moment conditions
• scientific article; zbMATH DE number 5811430 (Why is no real title available?)
• Model robust hybrid likelihood
• Bayesian jackknife empirical likelihood for the error variance in linear regression models
• Empirical likelihood based Bayesian variable selection
• Bayesian penalized empirical likelihood and Markov chain Monte Carlo sampling
• Gibbs posterior concentration rates under sub-exponential type losses
• Approximating Bayes in the 21st century
• Regularization of Bayesian quasi-likelihoods constructed from complex estimating functions
• Classical \(p\)-values and the Bayesian posterior probability that the hypothesis is approximately true
• Bernstein-von Mises theorem and misspecified models: a review

This page was built for publication: Bayesian Estimation and Comparison of Moment Condition Models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3121558)