On Limiting Distribution of Quasi-Posteriors under Partial Identification

TL;DR

This paper derives the limiting distribution of quasi-posteriors in models where parameters are only partially identified, providing theoretical insights into their asymptotic behavior.

Contribution

It establishes the total variation convergence of quasi-posteriors under partial identification, extending existing Bayesian asymptotic theory.

Findings

Limiting distribution characterized in total variation

Applicable to models with moment conditions

Provides theoretical foundation for quasi-posterior analysis

Abstract

We establish the limiting distribution (in total variation) of the quasi posteriors based on moment conditions, which only partially identify the parameters of interest. Some examples are discussed.