Performance Evaluation of Two Tests for the Equality of High-Dimensional Mean Vectors with Unequal Covariance Matrices
Main Article Content
Abstract
The rapid development of measurement technology in fields such as genetics, medicine, and economics has made high-dimensional data analysis an essential practice, as analysts and researchers need to collect and analyze data from a large number of variables. Analyzing high-dimensional data requires advanced statistical techniques as conventional methods like comparing mean vectors using Hotelling's T2-test are not applicable. In this study, we compare the performance of two tests, TMCQ and TSKK, for testing the equality of high-dimensional mean vectors when population covariance matrices are not equal. The TMCQ was initially proposed by Chen, S. X. and Qin, Y. L. in 2010 and modified by Srivastava, M. S., Katayama, S., & Kano, Y. in 2013, while the TSKK was developed by Srivastava, M. S. et al. in 2013. A simulation study was conducted using two independent normal samples of equal size. The population covariance matrices were created under five structures: Sphericity, Compound Symmetry (CS), Heterogeneous Compound Symmetry (CSH), Toeplitz, and Block Diagonal (BD) matrix. The results showed that the choice of covariance structure had an impact on the performance of both tests. The TMCQ test performed well for certain covariance structures including Sphericity, Toeplitz, and BD, while the TSKK performed well for the CS covariance structure. The performance of TSKK increased for large sample sizes of at least 60 under the covariance structures of Toeplitz and BD. Under the covariance structures of Sphericity, Toeplitz, and BD, the TMCQ outperformed the TSKK, while TSKK was more efficient than TMCQ when the covariance matrix structure was CS. Additionally, the performance of TSKK improved when the sample size was at least 60. We also examined the performance of the two tests when the difference between two population covariance matrices increased and found that both tests continued to perform well for certain covariance structures. However, under the covariance structure of CSH, both tests underperformed for the case being studied
Downloads
Metrics
Article Details
You are free to:
- Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices:
You do not have to comply with the license for elements of the material in the public domain or where your use is permitted by an applicable exception or limitation .
No warranties are given. The license may not give you all of the permissions necessary for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.