Adaptive Bayesian estimation using a Gaussian random field with inverse Gamma bandwidth

A.W. van der Vaart, J.H. van van Zanten

Research output: Contribution to JournalArticleAcademicpeer-review

115 Downloads (Pure)

Abstract

We Consider Nonparametric Bayesian Estimation Inference Using A Rescaled Smooth Gaussian Fld. As A Prior for A Multidimensional Funct.. the Rescaling Is Achieved Using A Gamma Variable and the Procedure Can Be Viewed As Choosing An Inverse Gamma Bandwidth. the Procedure Is Studied from A Frequentist Perspective in Three Stat. Settings Involving Replicated Observations . We Prove That the Resulting Posterior Distr. Shrinks to the Distr. That Generates the Data at A Speed Which Is Minimax-optimal Up to A Logarithmic Factor, Whatever the Regularity Level of the Data-generating Distr.. Thus the Hierachical Bayesian Procedure, with A Fixed Prior Is Shown to Be Fully Adaptive. Inst. of Math. Stat., 2009.
Original languageEnglish
Pages (from-to)2655-2675
Number of pages21
JournalAnnals of Statistics
Volume37
Issue number5B
DOIs
Publication statusPublished - 2009

Fingerprint

Dive into the research topics of 'Adaptive Bayesian estimation using a Gaussian random field with inverse Gamma bandwidth'. Together they form a unique fingerprint.

Cite this