Bayesian Analysis in Moment Inequality Models

This paper presents a study of the large-sample behavior of the posterior distribution of a structural parameter which is partially identified by moment inequalities. The posterior density is derived based on the limited information likelihood. The posterior distribution converges to zero exponentially fast on any δ -contraction outside the identified region. Inside, it is bounded below by a positive constant if the identified region is assumed to have a nonempty interior. Our simulation evidence indicates that the Bayesian approach has advantages over frequentist methods, in the sense that, with a proper choice of the prior, the posterior provides more information about the true parameter inside the identified region.We also address the problem of moment and model selection. Our optimality criterion is the maximum posterior procedure and we show that, asymptotically, it selects the true moment/model combination with the most moment inequalities and the simplest model.

[1]  Charles F. Manski,et al.  Confidence Intervals for Partially Identified Parameters , 2003 .

[2]  Francisco J. Samaniego,et al.  On the Efficacy of Bayesian Inference for Nonidentifiable Models , 1997 .

[3]  D. Andrews,et al.  Inference for Parameters Defined by Moment Inequalities: A Recommended Moment Selection Procedure , 2008 .

[4]  Federico A. Bugni.  Bootstrap Inference in Partially Identi…ed Models , 2009 .

[5]  V. Chernozhukov,et al.  Estimation and Confidence Regions for Parameter Sets in Econometric Models , 2007 .

[6]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[7]  P. Gustafson.  On Model Expansion, Model Contraction, Identifiability and Prior Information: Two Illustrative Scenarios Involving Mismeasured Variables , 2005 .

[8]  Y. Shao,et al.  Asymptotics for likelihood ratio tests under loss of identifiability , 2003 .

[9]  A. Gelfand,et al.  Identifiability, Improper Priors, and Gibbs Sampling for Generalized Linear Models , 1999 .

[10]  Jae-Young Kim,et al.  Limited information likelihood and Bayesian analysis , 2002 .

[11]  Francesca Molinari,et al.  Asymptotic Properties for a Class of Partially Identified Models , 2006 .

[12]  Adam M. Rosen,et al.  Confidence Sets for Partially Identified Parameters that Satisfy a Finite Number of Moment Inequalities , 2006 .

[13]  Frank Schorfheide,et al.  Bayesian and Frequentist Inference in Partially Identified Models , 2009 .

[14]  C. Manski,et al.  Inference on Regressions with Interval Data on a Regressor or Outcome , 2002 .

[15]  Inference for Parameters Defined by Moment Inequalities: A Recommended Moment Selection Procedure , 2008 .

[16]  Dale J. Poirier,et al.  REVISING BELIEFS IN NONIDENTIFIED MODELS , 1998, Econometric Theory.

[17]  Ivan A. Canay.  EL inference for partially identified models: Large deviations optimality and bootstrap validity , 2010 .

[18]  Azeem M. Shaikh,et al.  Inference for identifiable parameters in partially identified econometric models , 2006 .

[19]  Arnold Zellner,et al.  Models, prior information, and Bayesian analysis☆ , 1996 .

[20]  J. Horowitz,et al.  Nonparametric Analysis of Randomized Experiments with Missing Covariate and Outcome Data , 2000 .

[21]  D. Andrews,et al.  Inference for Parameters Defined by Moment Inequalities Using Generalized Moment Selection , 2007 .