Quasi-Bayesian analysis of nonparametric instrumental variables models

Abstract

This paper aims at developing a quasi-Bayesian analysis of the nonparametric instrumental variables model, with a focus on the asymptotic properties of quasi-posterior distributions. In this paper, instead of assuming a distributional assumption on the data generating process, we consider a quasi-likelihood induced from the conditional moment restriction, and put priors on the function-valued parameter. We call the resulting posterior quasi-posterior, which corresponds to ``Gibbs posterior'' in the literature. Here we focus on priors constructed on slowly growing finite-dimensional sieves. We derive rates of contraction and a nonparametric Bernstein-von Mises type result for the quasi-posterior distribution, and rates of convergence for the quasi-Bayes estimator defined by the posterior expectation. We show that, with priors suitably chosen, the quasi-posterior distribution (the quasi-Bayes estimator) attains the minimax optimal rate of contraction (convergence, resp.). These results greatly sharpen the previous related work.

Published in at http://dx.doi.org/10.1214/13-AOS1150 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (12)

• An MCMC Approach to Classical Estimation
• Nonparametric methods for inference in the presence of instrumental variables
• Convergence rates of posterior distributions for noniid observations
• From $\epsilon$-entropy to KL-entropy: Analysis of minimum information complexity density estimation
• Misspecification in infinite-dimensional Bayesian statistics
• Gibbs posterior for variable selection in high-dimensional classification and data mining
• Posterior Contraction Rates for the Bayesian Approach to Linear Ill-Posed Inverse Problems
• Quasi-Bayesian analysis of nonparametric instrumental variables models
• Bayesian recovery of the initial condition for the heat equation
• A Bernstein-Von Mises Theorem for discrete probability distributions
• Convergence rates for Bayesian density estimation of infinite-dimensional exponential families
• Bayes procedures for adaptive inference in inverse problems for the white noise model

Cited by in corpus (15)

• Optimal Uniform Convergence Rates and Asymptotic Normality for Series Estimators Under Weak Dependence and Weak Conditions
• On the Bernstein-von Mises phenomenon for nonparametric Bayes procedures
• Optimal Sup-norm Rates and Uniform Inference on Nonlinear Functionals of Nonparametric IV Regression
• Quasi-Bayesian analysis of nonparametric instrumental variables models
• On Oracle Property and Asymptotic Validity of Bayesian Generalized Method of Moments
• A Projection Framework for Testing Shape Restrictions That Form Convex Cones
• A Moreau-Yosida approximation scheme for a class of high-dimensional posterior distributions
• An approach to large-scale Quasi-Bayesian inference with spike-and-slab priors
• On frequentist coverage errors of Bayesian credible sets in moderately high dimensions
• A scalable quasi-Bayesian framework for Gaussian graphical models
• Bayesian variable selection in linear regression models with instrumental variables
• Bayesian Analysis of High-dimensional Discrete Graphical Models
• Some New Asymptotic Theory for Least Squares Series: Pointwise and Uniform Results
• Quasi-Bayesian Dual Instrumental Variable Regression
• Bayesian causal inference with some invalid instrumental variables