A combined adaptive-mixtures/plug-in estimator of multivariate probability densities

A multivariate extension of the plug-in kernel (and filtered kernel) estimator is proposed and this uses asymptotically optimal bandwidth matrix (matrices) for a normal mixture approximation of a density to be estimated (the filtered kernel estimator uses different matrices for different clusters of...

Full description

Saved in:

Bibliographic Details
Published in:	Computational statistics & data analysis Vol. 26; no. 2; pp. 199 - 218
Main Authors:	Ćwik, J, Koronacki, J
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 04.12.1997 Elsevier
Series:	Computational Statistics & Data Analysis
Subjects:	Gaussian clustering algorithm Nonparametric density estimation Plug-in kernel estimation Recursive EM algorithm Gaussian clustering algorithm Plug-in kernel estimation Nonparametric density estimation Recursive EM algorithm
ISSN:	0167-9473, 1872-7352
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	A multivariate extension of the plug-in kernel (and filtered kernel) estimator is proposed and this uses asymptotically optimal bandwidth matrix (matrices) for a normal mixture approximation of a density to be estimated (the filtered kernel estimator uses different matrices for different clusters of data). The normal mixture approximation is provided by a recursive version of the EM algorithm whose initial conditions are in turn obtained via an application of the ideas of adaptive mixtures density estimation and AIC-based pruning. Simulations show that the estimator proposed, while it is in fact a rather complex multistage estimation process, provides a very reliable way of estimating arbitrary and highly structured continuous densities on R 2 and, hopefully, R 3
Bibliography:	ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23
ISSN:	0167-9473 1872-7352
DOI:	10.1016/S0167-9473(97)00032-7