We study location-scale mixture priors for nonparametric statistical problems, including multivariate regression, density estimation and classification. We show that a rate-adaptive procedure can be obtained if the prior is properly constructed. In particular, we show that adaptation is achieved if a kernel mixture prior on a regression function is constructed using a Gaussian kernel, an inverse gamma bandwidth, and Gaussian mixing weights.
@article{1284988407,
author = {de Jonge, R. and van Zanten, J. H.},
title = {Adaptive nonparametric Bayesian inference using location-scale mixture priors},
journal = {Ann. Statist.},
volume = {38},
number = {1},
year = {2010},
pages = { 3300-3320},
language = {en},
url = {http://dml.mathdoc.fr/item/1284988407}
}
de Jonge, R.; van Zanten, J. H. Adaptive nonparametric Bayesian inference using location-scale mixture priors. Ann. Statist., Tome 38 (2010) no. 1, pp. 3300-3320. http://gdmltest.u-ga.fr/item/1284988407/