Calibrating general posterior credible regions
Authors:
- 10 December 2018
- journal article
- research article
- Published by Oxford University Press (OUP) in Biometrika
- Vol. 106 (2) , 479-486
Abstract
Summary: 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.Keywords
All Related Versions
Funding Information
- Army Research Offices
This publication has 20 references indexed in Scilit:
- Posterior Consistency of Bayesian Quantile Regression Based on the Misspecified Asymmetric Laplace DensityBayesian Analysis, 2013
- Risk of Bayesian Inference in Misspecified Models, and the Sandwich Covariance MatrixEconometrica, 2013
- The Bernstein-Von-Mises theorem under misspecificationElectronic Journal of Statistics, 2012
- Gibbs posterior for variable selection in high-dimensional classification and data miningThe Annals of Statistics, 2008
- From ɛ-entropy to KL-entropy: Analysis of minimum information complexity density estimationThe Annals of Statistics, 2006
- An MCMC approach to classical estimationJournal of Econometrics, 2003
- Bayesian empirical likelihoodBiometrika, 2003
- Bayesian quantile regressionStatistics & Probability Letters, 2001
- Bootstrap Methods for Median Regression ModelsEconometrica, 1998
- A Stochastic Approximation MethodThe Annals of Mathematical Statistics, 1951