Bayesian estimation and comparison of conditional moment models

We provide a Bayesian analysis of models in which the unknown distribution of the outcomes is speci?ed up to a set of conditional moment restrictions. This analysis is based on the nonparametric exponentially tilted empirical likelihood (ETEL) function, which is constructed to satisfy a sequence of unconditional moments, obtained from the conditional moments by an increasing (in sample size) vector of approximating functions (such as tensor splines based on the splines of each conditioning variable). The posterior distribution is shown to satisfy the Bernstein-von Mises theorem, subject to a growth rate condition on the number of approximating functions, even under misspeci?cation of the conditional moments. A large-sample theory for comparing different conditional moment models is also developed. The central result is that the marginal likelihood criterion selects the model that is less misspeci?ed, that is, the model that is closer to the unknown true distribution in terms of the Kullback-Leibler divergence. Several examples are provided to illustrate the framework and results.

[1]  Wenxin Jiang,et al.  Posterior Consistency of Nonparametric Conditional Moment Restricted Models , 2010, 1105.4847.

[2]  W. Newey,et al.  16 Efficient estimation of models with conditional moment restrictions , 1993 .

[3]  Van Der Vaart,et al.  UvA-DARE ( Digital Academic Repository ) The Bernstein-Von-Mises theorem under misspecification , 2012 .

[4]  P. Bickel,et al.  The semiparametric Bernstein-von Mises theorem , 2010, 1007.0179.

[5]  Stephen G. Donald,et al.  Choosing the Number of Instruments , 2001 .

[6]  G. Chamberlain.  Asymptotic efficiency in estimation with conditional moment restrictions , 1987 .

[7]  M. Hristache,et al.  Conditional moment models with data missing at random , 2017 .

[8]  W. Newey,et al.  Convergence rates and asymptotic normality for series estimators , 1997 .

[9]  Anna Simoni,et al.  Online appendix to : Bayesian Estimation and Comparison of Moment Condition Models , 2017 .

[10]  Anna Simoni,et al.  Gaussian Processes and Bayesian Moment Estimation , 2016, Journal of Business & Economic Statistics.

[11]  Yuichi Kitamura,et al.  Empirical Likelihood Based Inference in Conditional Moment Restriction Models , 2004 .

[12]  Posterior Consistency of Nonparametric Conditional Moment Restricted Models , 2010 .

[13]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[14]  Xiaohong Chen,et al.  Efficient Estimation of Models with Conditional Moment Restrictions Containing Unknown Functions , 2003 .

[15]  Kengo Kato,et al.  Quasi-Bayesian analysis of nonparametric instrumental variables models , 2012, 1204.2108.

[16]  H. White.  Maximum Likelihood Estimation of Misspecified Models , 1982 .

[17]  J. Florens,et al.  GENERALIZATION OF GMM TO A CONTINUUM OF MOMENT CONDITIONS , 2000, Econometric Theory.

[18]  Anna Simoni,et al.  REGULARIZING PRIORS FOR LINEAR INVERSE PROBLEMS , 2014, Econometric Theory.

[19]  S. Chib,et al.  Understanding the Metropolis-Hastings Algorithm , 1995 .

[20]  Elie Tamer,et al.  Monte Carlo Confidence Sets for Identified Sets , 2016, 1605.00499.

[21]  Stephen G. Donald,et al.  Choosing instrumental variables in conditional moment restriction models , 2009 .

[22]  S. Chib.  Marginal Likelihood from the Gibbs Output , 1995 .

[23]  Identification problem of the exponential tilting estimator under misspecification , 2013 .

[24]  Guido W. Imbens,et al.  Empirical likelihood estimation and consistent tests with conditional moment restrictions , 2003 .

[25]  I. Csiszár.  Sanov Property, Generalized $I$-Projection and a Conditional Limit Theorem , 1984 .

[26]  Chunrong Ai,et al.  Estimation of possibly misspecified semiparametric conditional moment restriction models with different conditioning variables , 2007 .

[27]  Anna Simoni,et al.  Nonparametric Estimation of An Instrumental Regression: A Quasi-Bayesian Approach Based on Regularized Posterior , 2012 .

[28]  Van Der Vaart,et al.  The Bernstein-Von-Mises theorem under misspecification , 2012 .

[29]  Susanne M. Schennach,et al.  Bayesian exponentially tilted empirical likelihood , 2005 .

[30]  S. Chib,et al.  Marginal Likelihood From the Metropolis–Hastings Output , 2001 .

[31]  S. Chib,et al.  Additive cubic spline regression with Dirichlet process mixture errors , 2010 .

[32]  Yuan Liao,et al.  Bayesian inference for partially identified smooth convex models , 2019, Journal of Econometrics.

[33]  Siddhartha Chib,et al.  Tailored randomized block MCMC methods with application to DSGE models , 2010 .

[34]  H. Bierens.  Consistent model specification tests , 1982 .

[35]  Gautam Tripathi,et al.  A matrix extension of the Cauchy-Schwarz inequality , 1999 .

[36]  S. Chib,et al.  Which Factors are Risk Factors in Asset Pricing? A Model Scan Framework , 2020, Journal of Business & Economic Statistics.