CriSP: A tool for bump hunting

Jaroslaw Harezlak, Nancy E. Heckman

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

We 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.

Original languageEnglish (US)
Pages (from-to)713-729
Number of pages17
JournalJournal of Computational and Graphical Statistics
Volume10
Issue number4
StatePublished - 2001
Externally publishedYes

Fingerprint

Smoothing Parameter
Regression Function
Derivatives
Probability density function
Derivative
Multimodality
Bootstrapping
Statistics
p-Value
Null hypothesis
Density Function
Test Statistic
Smoothing
Testing
Simulation Study
Necessary
Hunting

Keywords

  • Bootstrap
  • Critical smoothing parameter
  • Growth data
  • Multimodality testing
  • Smoothing

ASJC Scopus subject areas

  • Mathematics(all)
  • Computational Mathematics
  • Statistics and Probability

Cite this

CriSP : A tool for bump hunting. / Harezlak, Jaroslaw; Heckman, Nancy E.

In: Journal of Computational and Graphical Statistics, Vol. 10, No. 4, 2001, p. 713-729.

Research output: Contribution to journalArticle

Harezlak, Jaroslaw ; Heckman, Nancy E. / CriSP : A tool for bump hunting. In: Journal of Computational and Graphical Statistics. 2001 ; Vol. 10, No. 4. pp. 713-729.
@article{837c6690f97149d0bca719587e2f472b,
title = "CriSP: A tool for bump hunting",
abstract = "We 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.",
keywords = "Bootstrap, Critical smoothing parameter, Growth data, Multimodality testing, Smoothing",
author = "Jaroslaw Harezlak and Heckman, {Nancy E.}",
year = "2001",
language = "English (US)",
volume = "10",
pages = "713--729",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "American Statistical Association",
number = "4",

}

TY - JOUR

T1 - CriSP

T2 - A tool for bump hunting

AU - Harezlak, Jaroslaw

AU - Heckman, Nancy E.

PY - 2001

Y1 - 2001

N2 - We 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.

AB - We 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.

KW - Bootstrap

KW - Critical smoothing parameter

KW - Growth data

KW - Multimodality testing

KW - Smoothing

UR - http://www.scopus.com/inward/record.url?scp=0035563154&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035563154&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0035563154

VL - 10

SP - 713

EP - 729

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

IS - 4

ER -