A biascorrection for crossvalidation bandwidth selection when a kernel
estimate is based on dependent data
Martin Sköld
Department of Mathematical Statistics,
Lund Institute of Technology,
Lund University,
1998
ISSN 02811944
ISRN LUTFD2/NFMS3202SE

Abstract:

Leastsquares crossvalidation bandwidth selection for kernel density estimation
(Rudemo (1982), Bowman(1984)) has been shown to underestimate the optimal
bandwidth if data are positively correlated (Hart & Vieu (1990), Sköld
(1998)). We calculate the asymptotic bias for the LSCV criterion under a
continuoustime model and apply it as a correction term to discretetime
data that can be modeled as a smooth continuoustime process sampled at high
rate.


Key words:

nonparametric density estimation, kernel estimate, dependent data, bandwidth
selection, crossvalidation