File Download
  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Substitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing

TitleSubstitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing
Authors
KeywordsCLT for linear spectral statistics
Unbiased sample covariance matrix
Substitution principle
Testing on high-dimensional covariance matrix
High-dimensional sample covariance matrix
Large Fisher matrix
High-dimensional data
Issue Date2015
PublisherInstitute of Mathematical Statistics. The Journal's web site is located at https://imstat.org/journals-and-publications/annals-of-statistics/
Citation
The Annals of Statistics, 2015, v. 43 n. 2, p. 546-591 How to Cite?
AbstractSample covariance matrices are widely used in multivariate statistical analysis. The central limit theorems (CLTs) for linear spectral statistics of high-dimensional noncentralized sample covariance matrices have received considerable attention in random matrix theory and have been applied to many high-dimensional statistical problems. However, known population mean vectors are assumed for noncentralized sample covariance matrices, some of which even assume Gaussian-like moment conditions. In fact, there are still another two most frequently used sample covariance matrices: the ME (moment estimator, constructed by subtracting the sample mean vector from each sample vector) and the unbiased sample covariance matrix (by changing the denominator n as N=n−1 in the ME) without depending on unknown population mean vectors. In this paper, we not only establish the new CLTs for noncentralized sample covariance matrices when the Gaussian-like moment conditions do not hold but also characterize the nonnegligible differences among the CLTs for the three classes of high-dimensional sample covariance matrices by establishing a substitution principle: by substituting the adjusted sample size N=n−1 for the actual sample size n in the centering term of the new CLTs, we obtain the CLT of the unbiased sample covariance matrices. Moreover, it is found that the difference between the CLTs for the ME and unbiased sample covariance matrix is nonnegligible in the centering term although the only difference between two sample covariance matrices is a normalization by n and n−1, respectively. The new results are applied to two testing problems for high-dimensional covariance matrices.
Persistent Identifierhttp://hdl.handle.net/10722/217228
ISSN
2020 Impact Factor: 4.028
2015 SCImago Journal Rankings: 6.653
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZheng, S-
dc.contributor.authorBai, Z-
dc.contributor.authorYao, J-
dc.date.accessioned2015-09-18T05:52:50Z-
dc.date.available2015-09-18T05:52:50Z-
dc.date.issued2015-
dc.identifier.citationThe Annals of Statistics, 2015, v. 43 n. 2, p. 546-591-
dc.identifier.issn0090-5364-
dc.identifier.urihttp://hdl.handle.net/10722/217228-
dc.description.abstractSample covariance matrices are widely used in multivariate statistical analysis. The central limit theorems (CLTs) for linear spectral statistics of high-dimensional noncentralized sample covariance matrices have received considerable attention in random matrix theory and have been applied to many high-dimensional statistical problems. However, known population mean vectors are assumed for noncentralized sample covariance matrices, some of which even assume Gaussian-like moment conditions. In fact, there are still another two most frequently used sample covariance matrices: the ME (moment estimator, constructed by subtracting the sample mean vector from each sample vector) and the unbiased sample covariance matrix (by changing the denominator n as N=n−1 in the ME) without depending on unknown population mean vectors. In this paper, we not only establish the new CLTs for noncentralized sample covariance matrices when the Gaussian-like moment conditions do not hold but also characterize the nonnegligible differences among the CLTs for the three classes of high-dimensional sample covariance matrices by establishing a substitution principle: by substituting the adjusted sample size N=n−1 for the actual sample size n in the centering term of the new CLTs, we obtain the CLT of the unbiased sample covariance matrices. Moreover, it is found that the difference between the CLTs for the ME and unbiased sample covariance matrix is nonnegligible in the centering term although the only difference between two sample covariance matrices is a normalization by n and n−1, respectively. The new results are applied to two testing problems for high-dimensional covariance matrices.-
dc.languageeng-
dc.publisherInstitute of Mathematical Statistics. The Journal's web site is located at https://imstat.org/journals-and-publications/annals-of-statistics/-
dc.relation.ispartofThe Annals of Statistics-
dc.rights© Institute of Mathematical Statistics, 2015. This article is available online at https://doi.org/10.1214/14-AOS1292-
dc.subjectCLT for linear spectral statistics-
dc.subjectUnbiased sample covariance matrix-
dc.subjectSubstitution principle-
dc.subjectTesting on high-dimensional covariance matrix-
dc.subjectHigh-dimensional sample covariance matrix-
dc.subjectLarge Fisher matrix-
dc.subjectHigh-dimensional data-
dc.titleSubstitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing-
dc.typeArticle-
dc.identifier.emailYao, J: jeffyao@hku.hk-
dc.identifier.authorityYao, J=rp01473-
dc.description.naturepublished_or_final_version-
dc.identifier.doi10.1214/14-AOS1292-
dc.identifier.scopuseid_2-s2.0-84924959589-
dc.identifier.hkuros253939-
dc.identifier.volume43-
dc.identifier.issue2-
dc.identifier.spage546-
dc.identifier.epage591-
dc.identifier.isiWOS:000352757100004-
dc.publisher.placeUnited States-
dc.identifier.issnl0090-5364-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats