Title
Testing The Equality Of Correlation Matrices When Sample Correlation Matrices Are Dependent
Keywords
Elliptical distribution; Wald statistic
Abstract
We consider the situation in which a 2 m × 2 m correlation matrix is naturally partitioned into a 2 × 2 form, where each submatrix is m × m and we wish to test for the equality of the two correlation matrices appearing on the diagonal of the partitioned form. Standard tests for the equality of correlation matrices do not apply since the corresponding sample correlation matrices are correlated. We obtain Wald statistics under the assumption of multivariate normality as well as extensions that apply to elliptical distributions. © 2006 Elsevier B.V. All rights reserved.
Publication Date
6-1-2007
Publication Title
Journal of Statistical Planning and Inference
Volume
137
Issue
6
Number of Pages
1992-1997
Document Type
Article
Personal Identifier
scopus
DOI Link
https://doi.org/10.1016/j.jspi.2006.05.005
Copyright Status
Unknown
Socpus ID
33846353893 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/33846353893
STARS Citation
Schott, James R., "Testing The Equality Of Correlation Matrices When Sample Correlation Matrices Are Dependent" (2007). Scopus Export 2000s. 6579.
https://stars.library.ucf.edu/scopus2000/6579