A new test for multivariate normality and homoscedasticity

Douglas M. Hawkins

doi:10.1080/00401706.1981.10486244

A new test for multivariate normality and homoscedasticity

Douglas M. Hawkins

Statistics (Twin Cities)

Research output: Contribution to journal › Article › peer-review

84 Scopus citations

Abstract

Given data X_ijj = 1 to n_ii = 1 to g, one may wish to test whether the X_ij are normal with common covariance matrix. Letting X_i. denote the mean of the ith group and S the pooled covariance matrix, the quantities V_ij = (X_ij – X_i.)′S⁻¹(X_ij – X_i.) are used in the test. The exact null distribution of V_ij is noted, and the Anderson-Darling statistic is proposed for testing whether the data fit this null distribution. The new test has acceptable power, is flexible, and is easily implemented. © 1981 Taylor & Francis Group, LLC.

Original language	English (US)
Pages (from-to)	105-110
Number of pages	6
Journal	Technometrics
Volume	23
Issue number	1
DOIs	https://doi.org/10.1080/00401706.1981.10486244
State	Published - 1981

Keywords

Discriminant analysis
Homoscedasticity
Test for normality

Publisher link

10.1080/00401706.1981.10486244

Find It

Find It at U of M Libraries

Cite this

@article{37de00591c994678a88c34eb999eabfd,

title = "A new test for multivariate normality and homoscedasticity",

abstract = "Given data Xijj = 1 to nii = 1 to g, one may wish to test whether the Xij are normal with common covariance matrix. Letting Xi. denote the mean of the ith group and S the pooled covariance matrix, the quantities Vij = (Xij – Xi.)′S−1(Xij – Xi.) are used in the test. The exact null distribution of Vij is noted, and the Anderson-Darling statistic is proposed for testing whether the data fit this null distribution. The new test has acceptable power, is flexible, and is easily implemented. {\textcopyright} 1981 Taylor & Francis Group, LLC.",

keywords = "Discriminant analysis, Homoscedasticity, Test for normality",

author = "Hawkins, {Douglas M.}",

year = "1981",

doi = "10.1080/00401706.1981.10486244",

language = "English (US)",

volume = "23",

pages = "105--110",

journal = "Technometrics",

issn = "0040-1706",

publisher = "American Statistical Association",

number = "1",

}

TY - JOUR

T1 - A new test for multivariate normality and homoscedasticity

AU - Hawkins, Douglas M.

PY - 1981

Y1 - 1981

N2 - Given data Xijj = 1 to nii = 1 to g, one may wish to test whether the Xij are normal with common covariance matrix. Letting Xi. denote the mean of the ith group and S the pooled covariance matrix, the quantities Vij = (Xij – Xi.)′S−1(Xij – Xi.) are used in the test. The exact null distribution of Vij is noted, and the Anderson-Darling statistic is proposed for testing whether the data fit this null distribution. The new test has acceptable power, is flexible, and is easily implemented. © 1981 Taylor & Francis Group, LLC.

AB - Given data Xijj = 1 to nii = 1 to g, one may wish to test whether the Xij are normal with common covariance matrix. Letting Xi. denote the mean of the ith group and S the pooled covariance matrix, the quantities Vij = (Xij – Xi.)′S−1(Xij – Xi.) are used in the test. The exact null distribution of Vij is noted, and the Anderson-Darling statistic is proposed for testing whether the data fit this null distribution. The new test has acceptable power, is flexible, and is easily implemented. © 1981 Taylor & Francis Group, LLC.

KW - Discriminant analysis

KW - Homoscedasticity

KW - Test for normality

UR - http://www.scopus.com/inward/record.url?scp=0001934955&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001934955&partnerID=8YFLogxK

U2 - 10.1080/00401706.1981.10486244

DO - 10.1080/00401706.1981.10486244

M3 - Article

SN - 0040-1706

VL - 23

SP - 105

EP - 110

JO - Technometrics

JF - Technometrics

IS - 1

ER -

A new test for multivariate normality and homoscedasticity

Abstract

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this