File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Testing homogeneity of high-dimensional covariance matrices

TitleTesting homogeneity of high-dimensional covariance matrices
Authors
KeywordsAsymptotic normality
high-dimensional covariance matrix
homogeneity test
multi-sample comparison
power enhancement
Issue Date2020
PublisherAcademia Sinica, Institute of Statistical Science. The Journal's web site is located at http://www.stat.sinica.edu.tw/statistica/
Citation
Statistica Sinica, 2020, v. 30 n. 1, p. 35-53 How to Cite?
AbstractTesting the homogeneity of multiple high-dimensional covariance matrices is becoming increasingly critical in multivariate statistical analyses owing to the emergence of big data. Many existing homogeneity tests for covariance matrices focus on two populations, under specific situations, for example, either sparse or dense alternatives. As a result, these methods are not suitable for general cases that include multiple groups. We propose a power-enhancement high-dimensional test for multi-sample comparisons of covariance matrices, which includes homogeneity tests of two matrices as a special case. The proposed tests do not require a distributional assumption, and can handle both sparse and non-sparse structures. Based on random-matrix theory, the asymptotic normality properties of our tests are established under both the null and the alternative hypotheses. Numerical studies demonstrate the substantial gain in power for the proposed method. Furthermore, we illustrate the method using a gene expression data set from a breast cancer study.
Persistent Identifierhttp://hdl.handle.net/10722/288181
ISSN
2020 Impact Factor: 1.261
2015 SCImago Journal Rankings: 2.292
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZheng, S-
dc.contributor.authorLin, R-
dc.contributor.authorGuo, J-
dc.contributor.authorYin, G-
dc.date.accessioned2020-10-05T12:09:03Z-
dc.date.available2020-10-05T12:09:03Z-
dc.date.issued2020-
dc.identifier.citationStatistica Sinica, 2020, v. 30 n. 1, p. 35-53-
dc.identifier.issn1017-0405-
dc.identifier.urihttp://hdl.handle.net/10722/288181-
dc.description.abstractTesting the homogeneity of multiple high-dimensional covariance matrices is becoming increasingly critical in multivariate statistical analyses owing to the emergence of big data. Many existing homogeneity tests for covariance matrices focus on two populations, under specific situations, for example, either sparse or dense alternatives. As a result, these methods are not suitable for general cases that include multiple groups. We propose a power-enhancement high-dimensional test for multi-sample comparisons of covariance matrices, which includes homogeneity tests of two matrices as a special case. The proposed tests do not require a distributional assumption, and can handle both sparse and non-sparse structures. Based on random-matrix theory, the asymptotic normality properties of our tests are established under both the null and the alternative hypotheses. Numerical studies demonstrate the substantial gain in power for the proposed method. Furthermore, we illustrate the method using a gene expression data set from a breast cancer study.-
dc.languageeng-
dc.publisherAcademia Sinica, Institute of Statistical Science. The Journal's web site is located at http://www.stat.sinica.edu.tw/statistica/-
dc.relation.ispartofStatistica Sinica-
dc.subjectAsymptotic normality-
dc.subjecthigh-dimensional covariance matrix-
dc.subjecthomogeneity test-
dc.subjectmulti-sample comparison-
dc.subjectpower enhancement-
dc.titleTesting homogeneity of high-dimensional covariance matrices-
dc.typeArticle-
dc.identifier.emailYin, G: gyin@hku.hk-
dc.identifier.authorityYin, G=rp00831-
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.5705/ss.202017.0275-
dc.identifier.hkuros315659-
dc.identifier.volume30-
dc.identifier.issue1-
dc.identifier.spage35-
dc.identifier.epage53-
dc.identifier.isiWOS:000575675600003-
dc.publisher.placeTaiwan, Republic of China-
dc.identifier.issnl1017-0405-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats