Skip navigation
Skip navigation

Bandwidth choice for nonparametric classification

Hall, Peter; Kang, Kee-Hoon

Description

It is shown that, for kernel-based classification with univariate distributions and two populations, optimal bandwidth choice has a dichotomous character. If the two densities cross at just one point, where their curvatures have the same signs, then minimum Bayes risk is achieved using bandwidths which are an order of magnitude larger than those which minimize pointwise estimation error. On the other hand, if the curvature signs are different, or if there are multiple crossing points,...[Show more]

CollectionsANU Research Publications
Date published: 2005-04-25
Type: Journal article
URI: http://hdl.handle.net/1885/97938
Source: Annals of Statistics 2005, Vol. 33, No. 1, 284-306
DOI: 10.1214/009053604000000959

Download

File Description SizeFormat Image
01_Hall_Bandwidth_Choice_2005.pdf195.96 kBAdobe PDFThumbnail


Items in Open Research are protected by copyright, with all rights reserved, unless otherwise indicated.

Updated:  23 August 2018/ Responsible Officer:  University Librarian/ Page Contact:  Library Systems & Web Coordinator