Properties of the posterior distribution of a regression model based on Gaussian random fields

A. A. Zaitsev, E. V. Burnaev, V. G. Spokoiny

Research output: Contribution to journalArticlepeer-review

15 Citations (Scopus)

Abstract

We consider the regression problem based on Gaussian processes. We assume that the prior distribution on the vector of parameters in the corresponding model of the covariance function is non-informative. Under this assumption, we prove the Bernstein-von Mises theorem that states that the posterior distribution on the parameters vector is close to the corresponding normal distribution. We show results of numerical experiments that indicate that our results apply in practically important cases.

Original languageEnglish
Pages (from-to)1645-1655
Number of pages11
JournalAutomation and Remote Control
Volume74
Issue number10
DOIs
Publication statusPublished - Oct 2013
Externally publishedYes

Fingerprint

Dive into the research topics of 'Properties of the posterior distribution of a regression model based on Gaussian random fields'. Together they form a unique fingerprint.

Cite this