Weaker Conditions to Asymptotic Approximation to L_p-Norms

by Pablo Martínez-Camblor, Norberto Corral .

Abstract: We derive the asymptotic normality for $L_p$-norms of kernel density estimation under weaker conditions than those requested by Horvath (1991). In particular, the hypothesis (C5) proposed in the referenced paper restricts the useful measurements excluding the Lebesgue measurements to densities with infinite support. In this paper, this hypothesis is weakened and we guarantee the asymptotic normality to $\int|f_n(x)-f(x)|^pdx$\, for an ample variety of densities.

Key Words: Kernel Estimation, KMT process, Lp Norm

Pablo Martínez-Camblor, pmcamblor@hotmail.com
Norberto Corral, norbert@uniovi.es

Editor: Coronel-Brizio, Hector F., hcoronel@xal.megared.net.mx

READING THE ARTICLE: You can read the article in portable document (.pdf) format (232467 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 2156 times since JUNE 9, 2008.

Return to the InterStat Home Page.