Bayesian inference in partially identified semiparametric models

Faculty Member: 

Proposal Description:

This second project is to address the same partially identified semiparametric likelihood models. However, we will apply Bayesian approach to construct
frequentist confidence set for the identified set of parameters of interests in a semiparametric likelihood model. In particular, we will impose some flexible priors on sieve approximated likelihood based or moment based models. The proposal needs to run some new Monte Carlo studies to demonstrate the finite sample properties of the method.

Requisite Skills and Qualifications: Additional skills of knowing some Bayesian Monte Carlo simulation tools.

Project Type: 
Tobin RA
Project Type Year: 
Spring 2018 Tobin Research Projects