CriSP: A tool for bump hunting

Subscribe to email list

Please select the email list(s) to which you wish to subscribe.

You are here

CriSP: A tool for bump hunting

TitleCriSP: A tool for bump hunting
Publication TypeJournal Article
Year of Publication2001
AuthorsHarezlak, J, Heckman, NE
JournalJOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
Volume10
Pagination713-729
Date PublishedDEC
Type of ArticleArticle
ISSN1061-8600
Keywordsbootstrap, critical smoothing parameter, growth data, multimodality testing, Smoothing
AbstractWe propose a test of multimodality of regression functions and their derivatives. The test statistic is a critical smoothing parameter (CriSP), giving the minimum amount of smoothing necessary to force the regression function to satisfy the null hypothesis. The p values are computed via bootstrapping. Our idea is motivated by Silverman's test concerning the number of modes in the density function. Simulation studies indicate that the test works well, even when testing for bumps in the derivative. We apply CriSP to children's growth data, to study the number of spurts of growth.
DOI10.1198/106186001317243412