Rates of contraction of posterior distributions based on Gaussian process priors
Authors:
- A. W. van der VaartAVA. W. van der Vaart
- J. H. van ZantenJZJ. H. van Zanten
Open Access
- 1 June 2008
- journal article
- Published by Institute of Mathematical Statistics in The Annals of Statistics
- Vol. 36 (3) , 1435-1463
Abstract
We derive rates of contraction of posterior distributions on nonparametric or semiparametric models based on Gaussian processes. The rate of contraction is shown to depend on the position of the true parameter relative to the reproducing kernel Hilbert space of the Gaussian process and the small ball probabilities of the Gaussian process. We determine these quantities for a range of examples of Gaussian priors and in several statistical settings. For instance, we consider the rate of contraction of the posterior distribution based on sampling from a smooth density model when the prior models the log density as a (fractionally integrated) Brownian motion. We also consider regression with Gaussian errors and smooth classification under a logistic or probit link function combined with various priors.Keywords
All Related Versions
This publication has 30 references indexed in Scilit:
- Posterior convergence rates of Dirichlet mixtures at smooth densitiesThe Annals of Statistics, 2007
- Krein’s spectral theory and the Paley–Wiener expansion for fractional Brownian motionThe Annals of Probability, 2005
- A series expansion of fractional Brownian motionProbability Theory and Related Fields, 2004
- Convergence rates of posterior distributionsThe Annals of Statistics, 2000
- Bayesian Inference for Semiparametric Regression Using a Fourier RepresentationJournal of the Royal Statistical Society Series B: Statistical Methodology, 1999
- A Bayesian Approach to Robust Binary Nonparametric RegressionJournal of the American Statistical Association, 1998
- The Gaussian measure of shifted ballsProbability Theory and Related Fields, 1994
- Bayesian Analysis of Binary and Polychotomous Response DataJournal of the American Statistical Association, 1993
- Towards a practicable Bayesian nonparametric density estimatorBiometrika, 1991
- The Logistic Normal Distribution for Bayesian, Nonparametric, Predictive DensitiesJournal of the American Statistical Association, 1988