Bayesian model selection and the concentration of the posterior of hyperparameters

N. P. Baldin, V. G. Spokoiny

Research output: Contribution to journalArticlepeer-review

Abstract

The present paper offers a construction of a hyperprior that can be used for Bayesian model selection. This construction is inspired by the idea of the unbiased model selection in a penalized maximum likelihood approach. The main result shows a one-sided contraction of the posterior: the posterior mass is allocated on models of lower complexity than the oracle one.

Original languageEnglish
Pages (from-to)13-34
Number of pages22
JournalFundamental and Applied Mathematics
Volume18
Issue number2
Publication statusPublished - 2013
Externally publishedYes

Fingerprint

Dive into the research topics of 'Bayesian model selection and the concentration of the posterior of hyperparameters'. Together they form a unique fingerprint.

Cite this