High-dimensional testing for proportional covariance matrices

Koji Tsukuda, Shun Matsuura

Research output: Contribution to journalArticle

Abstract

Hypothesis testing for the proportionality of covariance matrices is a classical statistical problem and has been widely studied in the literature. However, there have been few treatments of this test in high-dimensional settings, especially for the case where the number of variables is larger than the sample size, despite high-dimensional statistical inference having recently received considerable attention. This paper studies hypothesis testing for the proportionality of two covariance matrices in the high-dimensional setting: m,n≍p δ for some δ∈(1∕2,1), where m and n denote the sample sizes and p denotes the number of variables. A test statistic is proposed and its asymptotic distribution is derived under multivariate normality. The non-asymptotic performance of the proposed test procedure is numerically examined.

Original languageEnglish
Pages (from-to)412-420
Number of pages9
JournalJournal of Multivariate Analysis
Volume171
DOIs
Publication statusPublished - 2019 May 1

Fingerprint

Covariance matrix
High-dimensional
Directly proportional
Hypothesis Testing
Testing
Sample Size
Denote
Multivariate Normality
Statistics
Statistical Inference
Asymptotic distribution
Test Statistic
Sample size
Hypothesis testing
Proportionality
Normality
Statistical inference
Test statistic

Keywords

  • Asymptotic test
  • High-dimension
  • Multivariate normal distribution
  • Proportional covariance model

ASJC Scopus subject areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

Cite this

High-dimensional testing for proportional covariance matrices. / Tsukuda, Koji; Matsuura, Shun.

In: Journal of Multivariate Analysis, Vol. 171, 01.05.2019, p. 412-420.

Research output: Contribution to journalArticle

@article{f3f79ede67d94a7695d8072122e19b14,
title = "High-dimensional testing for proportional covariance matrices",
abstract = "Hypothesis testing for the proportionality of covariance matrices is a classical statistical problem and has been widely studied in the literature. However, there have been few treatments of this test in high-dimensional settings, especially for the case where the number of variables is larger than the sample size, despite high-dimensional statistical inference having recently received considerable attention. This paper studies hypothesis testing for the proportionality of two covariance matrices in the high-dimensional setting: m,n≍p δ for some δ∈(1∕2,1), where m and n denote the sample sizes and p denotes the number of variables. A test statistic is proposed and its asymptotic distribution is derived under multivariate normality. The non-asymptotic performance of the proposed test procedure is numerically examined.",
keywords = "Asymptotic test, High-dimension, Multivariate normal distribution, Proportional covariance model",
author = "Koji Tsukuda and Shun Matsuura",
year = "2019",
month = "5",
day = "1",
doi = "10.1016/j.jmva.2019.01.011",
language = "English",
volume = "171",
pages = "412--420",
journal = "Journal of Multivariate Analysis",
issn = "0047-259X",
publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - High-dimensional testing for proportional covariance matrices

AU - Tsukuda, Koji

AU - Matsuura, Shun

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Hypothesis testing for the proportionality of covariance matrices is a classical statistical problem and has been widely studied in the literature. However, there have been few treatments of this test in high-dimensional settings, especially for the case where the number of variables is larger than the sample size, despite high-dimensional statistical inference having recently received considerable attention. This paper studies hypothesis testing for the proportionality of two covariance matrices in the high-dimensional setting: m,n≍p δ for some δ∈(1∕2,1), where m and n denote the sample sizes and p denotes the number of variables. A test statistic is proposed and its asymptotic distribution is derived under multivariate normality. The non-asymptotic performance of the proposed test procedure is numerically examined.

AB - Hypothesis testing for the proportionality of covariance matrices is a classical statistical problem and has been widely studied in the literature. However, there have been few treatments of this test in high-dimensional settings, especially for the case where the number of variables is larger than the sample size, despite high-dimensional statistical inference having recently received considerable attention. This paper studies hypothesis testing for the proportionality of two covariance matrices in the high-dimensional setting: m,n≍p δ for some δ∈(1∕2,1), where m and n denote the sample sizes and p denotes the number of variables. A test statistic is proposed and its asymptotic distribution is derived under multivariate normality. The non-asymptotic performance of the proposed test procedure is numerically examined.

KW - Asymptotic test

KW - High-dimension

KW - Multivariate normal distribution

KW - Proportional covariance model

UR - http://www.scopus.com/inward/record.url?scp=85061254094&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061254094&partnerID=8YFLogxK

U2 - 10.1016/j.jmva.2019.01.011

DO - 10.1016/j.jmva.2019.01.011

M3 - Article

AN - SCOPUS:85061254094

VL - 171

SP - 412

EP - 420

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

SN - 0047-259X

ER -