ADAPTIVE NONPARAMETRIC BAYESIAN INFERENCE USING LOCATION-SCALE MIXTURE PRIORS

作者：de Jonge R^*; van Zanten J H

来源：Annals of Statistics, 2010, 38(6): 3300-3320.

DOI：10.1214/10-AOS811

• Rate of convergence
• posterior distribution
• adaptation
• Bayesian inference
• nonparametric regression
• kernel mixture priors

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.

• 出版日期2010-12