Title
A Test For The Equality Of Covariance Matrices When The Dimension Is Large Relative To The Sample Sizes
Keywords
Equal covariance matrices; High-dimensional data; Singular sample covariance matrix
Abstract
A simple statistic is proposed for testing the equality of the covariance matrices of several multivariate normal populations. The asymptotic null distribution of this statistic, as both the sample sizes and the number of variables go to infinity, is shown to be normal. Consequently, this test can be used when the number of variables is not small relative to the sample sizes and, in particular, even when the number of variables exceeds the sample sizes. The finite sample size performance of the normal approximation for this method is evaluated in a simulation study. © 2007 Elsevier B.V. All rights reserved.
Publication Date
8-15-2007
Publication Title
Computational Statistics and Data Analysis
Volume
51
Issue
12
Number of Pages
6535-6542
Document Type
Article
Personal Identifier
scopus
DOI Link
https://doi.org/10.1016/j.csda.2007.03.004
Copyright Status
Unknown
Socpus ID
34547206225 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/34547206225
STARS Citation
Schott, James R., "A Test For The Equality Of Covariance Matrices When The Dimension Is Large Relative To The Sample Sizes" (2007). Scopus Export 2000s. 6430.
https://stars.library.ucf.edu/scopus2000/6430