Optimal Choice of the Kernel Function for the Parzen Kernel-type Density Estimators
Date
1975-04-20
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Let Wp(2) be the Sobolev space of probability density functions f(X) whose first derivative is absolutely continuous and whose second derivative is in Lp(- ∞ ,+ ∞), for p ∈ [1, + ∞]. Using an upper bound to the mean square error for a fixed X E [f(X) - fn(X)0]2, found by G. Wahba, where fn(X) is the Parzen Kernel-type estimate of f(X), we find the finite support Kernel function K(X) that minimizes the said upper bound. The optimal Kernel funciton is: K(y) = (1+a-1) (2T)-1 [1-T-a |y|a], for |y|<T where [-T,T] is the support interval, and a=2-p-1.
Description
Tech Report
Advisor
Degree
Type
Report
Keywords
kernel, parzen
Citation
D. Kazakos, "Optimal Choice of the Kernel Function for the Parzen Kernel-type Density Estimators," Rice University ECE Technical Report, no. TR7501, 1975.