An ANOVA-type nonparametric diagnostic test for heteroscedastic regression models

Lan Wang, Michael G. Akritas, Ingrid Van Keilegom

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

For the heteroscedastic nonparametric regression model Yni=m (xni)+σ (xnini, i=1,...,n, we discuss a novel method for testing some parametric assumptions about the regression function m. The test is motivated by recent developments in the asymptotic theory for analysis of variance when the number of factor levels is large. Asymptotic normality of the test statistic is established under the null hypothesis and suitable local alternatives. The similarity of the form of the test statistic to that of the classical F-statistic in analysis of variance allows easy and fast calculation. Simulation studies demonstrate that the new test possesses satisfactory finite-sample properties.

Original languageEnglish (US)
Pages (from-to)365-382
Number of pages18
JournalJournal of Nonparametric Statistics
Volume20
Issue number5
DOIs
StatePublished - Jul 1 2008

Fingerprint

Heteroscedastic Regression
Heteroscedastic Model
Diagnostic Tests
Non-parametric test
Analysis of variance
Test Statistic
Regression Model
F-statistics
Local Alternatives
Nonparametric Model
Asymptotic Theory
Nonparametric Regression
Regression Function
Asymptotic Normality
Null hypothesis
Simulation Study
Testing
Demonstrate
Regression model
Test statistic

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{92d3a2dd618c45ea815630ed067cc9a9,
title = "An ANOVA-type nonparametric diagnostic test for heteroscedastic regression models",
abstract = "For the heteroscedastic nonparametric regression model Yni=m (xni)+σ (xni)εni, i=1,...,n, we discuss a novel method for testing some parametric assumptions about the regression function m. The test is motivated by recent developments in the asymptotic theory for analysis of variance when the number of factor levels is large. Asymptotic normality of the test statistic is established under the null hypothesis and suitable local alternatives. The similarity of the form of the test statistic to that of the classical F-statistic in analysis of variance allows easy and fast calculation. Simulation studies demonstrate that the new test possesses satisfactory finite-sample properties.",
author = "Lan Wang and Akritas, {Michael G.} and {Van Keilegom}, Ingrid",
year = "2008",
month = "7",
day = "1",
doi = "10.1080/10485250802066112",
language = "English (US)",
volume = "20",
pages = "365--382",
journal = "Journal of Nonparametric Statistics",
issn = "1048-5252",
publisher = "Taylor and Francis Ltd.",
number = "5",

}

An ANOVA-type nonparametric diagnostic test for heteroscedastic regression models. / Wang, Lan; Akritas, Michael G.; Van Keilegom, Ingrid.

In: Journal of Nonparametric Statistics, Vol. 20, No. 5, 01.07.2008, p. 365-382.

Research output: Contribution to journalArticle

TY - JOUR

T1 - An ANOVA-type nonparametric diagnostic test for heteroscedastic regression models

AU - Wang, Lan

AU - Akritas, Michael G.

AU - Van Keilegom, Ingrid

PY - 2008/7/1

Y1 - 2008/7/1

N2 - For the heteroscedastic nonparametric regression model Yni=m (xni)+σ (xni)εni, i=1,...,n, we discuss a novel method for testing some parametric assumptions about the regression function m. The test is motivated by recent developments in the asymptotic theory for analysis of variance when the number of factor levels is large. Asymptotic normality of the test statistic is established under the null hypothesis and suitable local alternatives. The similarity of the form of the test statistic to that of the classical F-statistic in analysis of variance allows easy and fast calculation. Simulation studies demonstrate that the new test possesses satisfactory finite-sample properties.

AB - For the heteroscedastic nonparametric regression model Yni=m (xni)+σ (xni)εni, i=1,...,n, we discuss a novel method for testing some parametric assumptions about the regression function m. The test is motivated by recent developments in the asymptotic theory for analysis of variance when the number of factor levels is large. Asymptotic normality of the test statistic is established under the null hypothesis and suitable local alternatives. The similarity of the form of the test statistic to that of the classical F-statistic in analysis of variance allows easy and fast calculation. Simulation studies demonstrate that the new test possesses satisfactory finite-sample properties.

UR - http://www.scopus.com/inward/record.url?scp=47549098483&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47549098483&partnerID=8YFLogxK

U2 - 10.1080/10485250802066112

DO - 10.1080/10485250802066112

M3 - Article

AN - SCOPUS:47549098483

VL - 20

SP - 365

EP - 382

JO - Journal of Nonparametric Statistics

JF - Journal of Nonparametric Statistics

SN - 1048-5252

IS - 5

ER -