The Bernstein-von Mises theorem and nonregular models

DOI10.1214/14-AOS1239zbMATH Open1305.62112arXiv1211.3434OpenAlexW3102766412MaRDI QIDQ480970FDOQ480970

P. J. Green, Natalia Bochkina

Publication date: 12 December 2014

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We study the asymptotic behaviour of the posterior distribution in a broad class of statistical models where the "true" solution occurs on the boundary of the parameter space. We show that in this case Bayesian inference is consistent, and that the posterior distribution has not only Gaussian components as in the case of regular models (the Bernstein-von Mises theorem) but also has Gamma distribution components whose form depends on the behaviour of the prior distribution near the boundary and have a faster rate of convergence. We also demonstrate a remarkable property of Bayesian inference, that for some models, there appears to be no bound on efficiency of estimating the unknown parameter if it is on the boundary of the parameter space. We illustrate the results on a problem from emission tomography.

Full work available at URL: https://arxiv.org/abs/1211.3434

zbMATH Keywords

Bayesian inferenceboundaryposterior concentrationtomographytotal variation distanceBernstein-von Mises theoremnonregularSPECTapproximate posteriorvariance estimation in mixed models

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Bayesian inference (62F15)

Cited In (19)

• Weak convergence of posteriors conditional on maximum pseudo-likelihood estimates and implications in ABC
• Penalising model component complexity: a principled, practical approach to constructing priors
• Adaptive Bayesian inference for current status data on a grid
• Finite sample Bernstein-von Mises theorem for semiparametric problems
• Title not available (Why is that?)
• Dimension Free Nonasymptotic Bounds on the Accuracy of High-Dimensional Laplace Approximation
• Bayesian fixed-domain asymptotics for covariance parameters in a Gaussian process model
• How principled and practical are penalised complexity priors?
• Selection of KL neighbourhood in robust Bayesian inference
• Critical dimension in the semiparametric Bernstein-von Mises theorem
• Additive Bayesian variable selection under censoring and misspecification
• Posterior analysis of \(n\) in the binomial \((n,p)\) problem with both parameters unknown -- with applications to quantitative nanoscopy
• Nonparametric Posterior Learning for Emission Tomography
• Posterior distribution of nondifferentiable functions
• A reverse Gaussian correlation inequality by adding cones
• Posterior contraction rates for support boundary recovery
• Maximum-a-Posteriori Estimation with Bayesian Confidence Regions
• Bernstein-von Mises theorem and misspecified models: a review
• Bayesian inference for the weights in logarithmic pooling

This page was built for publication: The Bernstein-von Mises theorem and nonregular models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q480970)