TY  - JOUR
AU  - Moon,Hyungsik Roger
AU  - Schorfheide,Frank
TI  - Bayesian and Frequentist Inference in Partially Identified Models
JF  - National Bureau of Economic Research Working Paper Series
VL  - No. 14882
PY  - 2009
Y2  - April 2009
UR  - http://www.nber.org/papers/w14882
L1  - http://www.nber.org/papers/w14882.pdf
N1  - Author contact info:
Hyungsik Moon
University of Southern California
Department of Economics
KAP 300
University Park Campus
Los Angeles, CA 90089
E-Mail: hyungsikmoon@gmail.com

Frank Schorfheide
University of Pennsylvania
Department of Economics
3718 Locust Walk
McNeil 525
Philadelphia, PA 19104-6297
Tel: 215/898-8486
Fax: 215/573-2057
E-Mail: schorf@ssc.upenn.edu
AB  - A large sample approximation of the posterior distribution of partially identified structural parameters is derived for models that can be indexed by a finite-dimensional reduced form parameter vector. It is used to analyze the differences between frequentist confidence sets and Bayesian credible sets in partially identified models. A key difference is that frequentist set estimates extend beyond the boundaries of the identified set (conditional on the estimated reduced form parameter), whereas Bayesian credible sets can asymptotically be located in the interior of the identified set. Our asymptotic approximations are illustrated in the context of simple moment inequality models and a numerical illustration for a two-player entry game is provided.
ER  -