Electronic Journal of Statistics

Invariant test based on the modified correction to LRT for the equality of two high-dimensional covariance matrices

Qiuyan Zhang, Jiang Hu, and Zhidong Bai

Full-text: Open access

Enhanced PDF (2214 KB)

Abstract

In this paper, we propose an invariant test based on the modified correction to the likelihood ratio test (LRT) of the equality of two high-dimensional covariance matrices. It is well-known that the classical log-LRT is not well defined when the dimension is larger than or equal to one of the sample sizes. Or even the log-LRT is well-defined, it is usually perceived as a bad statistic in the high-dimensional cases because of their low powers under some alternatives. In this paper, we will justify the usefulness of the modified log-LRT, and an invariant test that works well in cases where the dimension is larger than the sample sizes. Besides, the test is established under the weakest conditions on the dimensions and the moments of the samples. The asymptotic distribution of the proposed test statistic is also obtained under the null hypothesis. What is more, we also propose a lite version of the modified LRT in the paper. A simulation study and a real data analysis show that the performances of the two proposed statistics are invariant under affine transformations.

Article information

Source
Electron. J. Statist., Volume 13, Number 1 (2019), 850-881.

Dates
Received: July 2018
First available in Project Euclid: 26 March 2019

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1553565706

Digital Object Identifier
doi:10.1214/19-EJS1542

Mathematical Reviews number (MathSciNet)
MR3932609

Zentralblatt MATH identifier
07056142

Subjects
Primary: 62H15: Hypothesis testing
Secondary: 62H10: Distribution of statistics

Keywords
Likelihood ratio test high-dimensional data hypothesis testing random matrix theory

Rights
Creative Commons Attribution 4.0 International License.

Citation

Zhang, Qiuyan; Hu, Jiang; Bai, Zhidong. Invariant test based on the modified correction to LRT for the equality of two high-dimensional covariance matrices. Electron. J. Statist. 13 (2019), no. 1, 850--881. doi:10.1214/19-EJS1542. https://projecteuclid.org/euclid.ejs/1553565706


Export citation

References

  • [1] Z. D. Bai, J. Hu, G. Pan, and W. Zhou. Convergence of the empirical spectral distribution function of Beta matrices., Bernoulli, 21(3) :1538–1574, 2015.
    Zentralblatt MATH: 1319.60006
    Digital Object Identifier: doi:10.3150/14-BEJ613
    Project Euclid: euclid.bj/1432732029
  • [2] Z. D. Bai and J. W. Silverstein. CLT for linear spectral statistics of large-dimensional sample covariance matrices., The Annals of Probability, 32(1):553–605, 2004.
    Zentralblatt MATH: 1063.60022
    Digital Object Identifier: doi:10.1214/aop/1078415845
    Project Euclid: euclid.aop/1078415845
  • [3] Z. D. Bai and J. W. Silverstein., Spectral analysis of large dimensional random matrices. Second Edition. Springer Verlag, 2010.
    Zentralblatt MATH: 1301.60002
  • [4] Z. D. Bai and Y. Q. Yin. Limit of the smallest eigenvalue of a large dimensional sample covariance matrix., The Annals of Probability, 21(3) :1275–1294, 1993.
    Zentralblatt MATH: 0779.60026
    Digital Object Identifier: doi:10.1214/aop/1176989118
    Project Euclid: euclid.aop/1176989118
  • [5] Z. D. Bai, D. D. Jiang, J. F. Yao, and S. R. Zheng. Corrections to LRT on large-dimensional covariance matrix by RMT., The Annals of Statistics, 37(6B) :3822–3840, 2009.
    Zentralblatt MATH: 1360.62286
    Digital Object Identifier: doi:10.1214/09-AOS694
    Project Euclid: euclid.aos/1256303528
  • [6] M. S. Bartlett. Properties of Sufficiency and Statistical Tests., Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 160(901):268–282, 1937.
    Zentralblatt MATH: 0016.41201
    Digital Object Identifier: doi:10.1098/rspa.1937.0109
  • [7] T. Cai, W. Liu, and Y. Xia. Two-Sample Covariance Matrix Testing and Support Recovery in High-Dimensional and Sparse Settings., Journal of the American Statistical Association, 108(501):265–277, 2013.
    Zentralblatt MATH: 06158341
    Digital Object Identifier: doi:10.1080/01621459.2012.758041
  • [8] T. T. Cai and Z. Ma. Optimal hypothesis testing for high dimensional covariance matrices., Bernoulli, 19(5B) :2359–2388, 2013.
    Zentralblatt MATH: 1281.62140
    Digital Object Identifier: doi:10.3150/12-BEJ455
    Project Euclid: euclid.bj/1386078606
  • [9] L. Chen, D. Paul, R. Prentice, and P. Wang. A regularized Hotelling’s T (2) test for pathway analysis in proteomic studies., Journal of the American Statistical Association, 106(496) :1345–1360, 2011.
    Zentralblatt MATH: 1234.62082
    Digital Object Identifier: doi:10.1198/jasa.2011.ap10599
  • [10] R. Cont. Empirical properties of asset returns: stylized facts and statistical issues., Quantitative Finance, 1:223–236, 2001.
    Zentralblatt MATH: 1408.62174
    Digital Object Identifier: doi:10.1080/713665670
  • [11] D. Jiang, T. Jiang, and F. Yang. Likelihood ratio tests for covariance matrices of high-dimensional normal distributions., Journal of Statistical Planning and Inference, 142(8) :2241–2256, 2012.
    Zentralblatt MATH: 1244.62082
    Digital Object Identifier: doi:10.1016/j.jspi.2012.02.057
  • [12] T. Jiang and F. Yang. Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions., The Annals of Statistics, 41(4) :2029–2074, 2013.
    Zentralblatt MATH: 1277.62149
    Digital Object Identifier: doi:10.1214/13-AOS1134
    Project Euclid: euclid.aos/1382547512
  • [13] O. Ledoit and M. Wolf. Some hypothesis tests for the covariance matrix when the dimension is large compared to the sample size., The Annals of Statistics, 30(4) :1081–1102, 2002.
    Zentralblatt MATH: 1029.62049
    Digital Object Identifier: doi:10.1214/aos/1031689018
    Project Euclid: euclid.aos/1031689018
  • [14] J. Li and S. X. Chen. Two sample tests for high-dimensional covariance matrices., The Annals of Statistics, 40(2):908–940, 2012.
    Zentralblatt MATH: 1274.62383
    Digital Object Identifier: doi:10.1214/12-AOS993
    Project Euclid: euclid.aos/1338515142
  • [15] G. Pan and W. Zhou. Central limit theorem for Hotelling’s $T^2$ statistic under large dimension., The Annals of Applied Probability, 21(5) :1860–1910, 2011.
    Zentralblatt MATH: 1250.62030
    Digital Object Identifier: doi:10.1214/10-AAP742
    Project Euclid: euclid.aoap/1319576611
  • [16] K. C. S. Pillai and K. Jayachandran. Power comparisons of tests of two multivariate hypotheses based on four criteria., Biometrika, 54(1):195–210, 1967.
  • [17] K. C. S. Pillai and K. Jayachandran. Power comparisons of tests of equality of two covariance matrices based on four criteria., Biometrika, 55(2):335–342, 1968.
  • [18] M. S. Srivastava and H. Yanagihara. Testing the equality of several covariance matrices with fewer observations than the dimension., Journal of Multivariate Analysis, 101(6) :1319–1329, 2010.
    Zentralblatt MATH: 1186.62078
    Digital Object Identifier: doi:10.1016/j.jmva.2009.12.010
  • [19] S. S. Wilks. Certain generalizations in the analysis of variance., Biometrika, 24(3):471–494, 1932.
    Zentralblatt MATH: 58.1172.02
    Digital Object Identifier: doi:10.1093/biomet/24.3-4.471
  • [20] S. S. Wilks. Sample criteria for testing equality of means, equality of variances, and equality of covariances in a normal multivariate distribution., The Annals of Mathematical Statistics, 257–281, 1946.
    Zentralblatt MATH: 0063.08259
    Digital Object Identifier: doi:10.1214/aoms/1177730940
    Project Euclid: euclid.aoms/1177730940
  • [21] S. Zheng, Z. Bai, and J. Yao. Substitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing., Annals of Statistics, 43(2):546–591, 2015.
    Zentralblatt MATH: 1312.62074
    Digital Object Identifier: doi:10.1214/14-AOS1292
    Project Euclid: euclid.aos/1424787428
  • [22] S. Zheng, Z. Bai, and J. Yao. CLT for eigenvalue statistics of large-dimensional general Fisher matrices with applications., Bernoulli, 23(2) :1130–1178, 2017.
    Zentralblatt MATH: 1375.60066
    Digital Object Identifier: doi:10.3150/15-BEJ772
    Project Euclid: euclid.bj/1486177395

The Institute of Mathematical Statistics and the Bernoulli Society

Electronic Journal of Statistics

New content alerts

Email RSS ToC RSS Article
  • You have access to this content.
  • You have partial access to this content.
  • You do not have access to this content.
Turn Off MathJax

What is MathJax?

More like this

+ See more