Open Access
April 2005 Default priors for Gaussian processes
Rui Paulo
Ann. Statist. 33(2): 556-582 (April 2005). DOI: 10.1214/009053604000001264


Motivated by the statistical evaluation of complex computer models, we deal with the issue of objective prior specification for the parameters of Gaussian processes. In particular, we derive the Jeffreys-rule, independence Jeffreys and reference priors for this situation, and prove that the resulting posterior distributions are proper under a quite general set of conditions. A proper flat prior strategy, based on maximum likelihood estimates, is also considered, and all priors are then compared on the grounds of the frequentist properties of the ensuing Bayesian procedures. Computational issues are also addressed in the paper, and we illustrate the proposed solutions by means of an example taken from the field of complex computer model validation.


Download Citation

Rui Paulo. "Default priors for Gaussian processes." Ann. Statist. 33 (2) 556 - 582, April 2005.


Published: April 2005
First available in Project Euclid: 26 May 2005

zbMATH: 1069.62030
MathSciNet: MR2163152
Digital Object Identifier: 10.1214/009053604000001264

Primary: 62F15
Secondary: 60G15 , 62M30

Keywords: computer model , frequentist coverage , Gaussian process , integrated likelihood , Jeffreys prior , posterior propriety , reference prior

Rights: Copyright © 2005 Institute of Mathematical Statistics

Vol.33 • No. 2 • April 2005
Back to Top