SFB 303 Discussion Paper No. A - 181

Author: Marron, J.S.
Title: Comments on a Data Based Bandwidth Selector
Abstract: A recently proposed data based method for choosing the bandwidth of a kernel density estimator is considered. Intuitive and asymptotic reasons are given for why the bandwidth selected in this fashion is smaller than it should be. This conflicts with the results of a simulation study. The conflict is resolved through a deeper asymptotic analysis. Further simulation results investigate the issue of what sample sizes are required for the asymptotics to properly describe the situation. The analysis is extended to motivate a type of correction, which leads to the widely known least-squares cross-validation, hence providing a new derivation of the latter.
Keywords: bandwidth selection, cross-validation, kernel density estimation, smoothing parameter
Creation-Date: July 1988
Unfortunately this paper is not available online. Please contact us to order a hardcopy.

SFB 303 Homepage

19.10.1999, Webmaster