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.
Original language  English 

Pages (fromto)  18501878 
Number of pages  29 
Journal  Annals of Statistics 
Volume  42 
DOIs  
Publication status  Published  2014 
Keywords
 Approximate posterior, Bayesian inference, Bernstein–von Mises theorem, boundary, nonregular, posterior concentration, SPECT, tomography, total variation distance, variance estimation in mixed models.
Fingerprint
Dive into the research topics of 'The Bernsteinvon Mises theorem and nonregular models'. Together they form a unique fingerprint.Profiles

Professor Peter J Green
 Statistical Science
 School of Mathematics  Professorial Research Fellow
 Cabot Institute for the Environment
 Statistics
Person: Academic , Member