Smoothness in Bayesian Non-parametric Regression with Wavelets |
| |
Authors: | Marco Di Zio Arnoldo Frigessi |
| |
Institution: | (1) ISTAT, Roma, Italy;(2) Norwegian Computing Center, P.O. Box 114, Blindern, Oslo, Norway |
| |
Abstract: | Consider the classical nonparametric regression problem yi = f(ti) + ii = 1,...,n where ti = i/n, and i are i.i.d. zero mean normal with variance 2. The aim is to estimate the true function f which is assumed to belong to the smoothness class described by the Besov space B
pq
q
. These are functions belonging to Lp with derivatives up to order s, in Lp sense. The parameter q controls a further finer degree of smoothness. In a Bayesian setting, a prior on B
pq
q
is chosen following Abramovich, Sapatinas and Silverman (1998). We show that the optimal Bayesian estimator of f is then also a.s. in B
pq
q
if the loss function is chosen to be the Besov norm of B
pq
q
. Because it is impossible to compute this optimal Bayesian estimator analytically, we propose a stochastic algorithm based on an approximation of the Bayesian risk and simulated annealing. Some simulations are presented to show that the algorithm performs well and that the new estimator is competitive when compared to the more standard posterior mean. |
| |
Keywords: | Besov space loss function mixture prior optimal Bayesian estimator simulated annealing |
本文献已被 SpringerLink 等数据库收录! |
|