Miscellanea Calibrating general posterior credible regions

• 成果类型: Article
• 署名作者: Syring, Nicholas; Martin, Ryan
• 署名单位: University of Illinois System; University of Illinois Chicago; University of Illinois Chicago Hospital; North Carolina State University
• 刊物名称: BIOMETRIKA
• ISSN/ISSBN: 0006-3444
• DOI: 10.1093/biomet/asy054
• 发表日期: 2019
• 页码: 479486
• 关键词: bayesian-inference; gibbs posterior; likelihood; models

摘要

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.