Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions
In this paper, a new test statistic based on the weighted Frobenius norm of covariance matrices is proposed to test the homogeneity of multi-group population covariance matrices. The asymptotic distributions of the proposed test under the null and the alternative hypotheses are derived, respectively...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/22/4339 |
| _version_ | 1797464636459581440 |
|---|---|
| author | Peng Sun Yincai Tang Mingxiang Cao |
| author_facet | Peng Sun Yincai Tang Mingxiang Cao |
| author_sort | Peng Sun |
| collection | DOAJ |
| description | In this paper, a new test statistic based on the weighted Frobenius norm of covariance matrices is proposed to test the homogeneity of multi-group population covariance matrices. The asymptotic distributions of the proposed test under the null and the alternative hypotheses are derived, respectively. Simulation results show that the proposed test procedure tends to outperform some existing test procedures. |
| first_indexed | 2024-03-09T18:11:03Z |
| format | Article |
| id | doaj.art-9b39f9255873483e90d651cc8102e211 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T18:11:03Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-9b39f9255873483e90d651cc8102e2112023-11-24T09:09:57ZengMDPI AGMathematics2227-73902022-11-011022433910.3390/math10224339Homogeneity Test of Multi-Sample Covariance Matrices in High DimensionsPeng Sun0Yincai Tang1Mingxiang Cao2Department of Statistics, East China Normal University, Shanghai 200062, ChinaDepartment of Statistics, East China Normal University, Shanghai 200062, ChinaSchool of Mathematics and Statistics, Anhui Normal University, Anhui 241002, ChinaIn this paper, a new test statistic based on the weighted Frobenius norm of covariance matrices is proposed to test the homogeneity of multi-group population covariance matrices. The asymptotic distributions of the proposed test under the null and the alternative hypotheses are derived, respectively. Simulation results show that the proposed test procedure tends to outperform some existing test procedures.https://www.mdpi.com/2227-7390/10/22/4339high-dimensional dataweighted Frobenius normhomogeneity testmartingale central limit theoremasymptotic distributions |
| spellingShingle | Peng Sun Yincai Tang Mingxiang Cao Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions Mathematics high-dimensional data weighted Frobenius norm homogeneity test martingale central limit theorem asymptotic distributions |
| title | Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions |
| title_full | Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions |
| title_fullStr | Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions |
| title_full_unstemmed | Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions |
| title_short | Homogeneity Test of Multi-Sample Covariance Matrices in High Dimensions |
| title_sort | homogeneity test of multi sample covariance matrices in high dimensions |
| topic | high-dimensional data weighted Frobenius norm homogeneity test martingale central limit theorem asymptotic distributions |
| url | https://www.mdpi.com/2227-7390/10/22/4339 |
| work_keys_str_mv | AT pengsun homogeneitytestofmultisamplecovariancematricesinhighdimensions AT yincaitang homogeneitytestofmultisamplecovariancematricesinhighdimensions AT mingxiangcao homogeneitytestofmultisamplecovariancematricesinhighdimensions |