Research work · 2014
Kernel Smoothing
Martin L. Hazelton
Wiley StatsRef: Statistics Reference Online
Abstract
Abstract Kernel smoothing is a nonparametric technique for exposing structure or trend in data. It can be applied in a wide variety of settings, including the problems of density estimation and nonparametric regression on which we focus attention. In each case, the performance of kernel methods is highly dependent on the choice of bandwidth, a parameter that controls the amount of smoothing applied to the data.