Calibrating general posterior credible regions
Abstract
Calibration of credible regions derived from under- or misspecified models is an important and challenging problem. In this paper, we introduce a scalar tuning parameter that controls the posterior distribution spread, and develop a Monte Carlo algorithm that sets this parameter so that the corresponding credible region achieves the nominal frequentist coverage probability.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Dec 10, 2018
- Source ID
- 10.1093/biomet/asy054
Entities
People
- Nicholas Syring
- Ryan Martin
Organizations
- Army Research Office
- North Carolina State University
- University of Illinois at Chicago