Study on the Overlap in Matrix-Variate Data with Applications in Discriminant Analysis

Yingying Zhang (), Volodymyr Melnykov () and Xuwen Zhu ()
Additional contact information
Yingying Zhang: Western Michigan University
Volodymyr Melnykov: The University of Alabama
Xuwen Zhu: The University of Alabama

Sankhya A: The Indian Journal of Statistics, 2025, vol. 87, issue 2, No 5, 378-403

Abstract: Abstract This paper introduces and quantifies the concept of overlap in matrix-variate data. The proposed methodology derives and computes the exact overlap, defined as the total misclassification probability between two clusters. Discriminant functions tailored for matrix-variate data are developed based on a specified level of overlap, and their effectiveness is evaluated. The performance of these classifiers is compared to that of traditional multivariate classifiers across various simulation scenarios, considering different overlap levels, sample sizes, and matrix dimensions. Results from both simulated and real-world datasets demonstrate the superior performance of discriminant functions based on the matrix distribution format.

Keywords: Discriminant classifiers; Maximum likelihood estimation (MLE); Matrix normal distribution; Matrix t distribution; Overlap character; User-specified algorithm design; 62H30; 62H10 (search for similar items in EconPapers)
Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s13171-025-00406-9 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sankha:v:87:y:2025:i:2:d:10.1007_s13171-025-00406-9

Ordering information: This journal article can be ordered from
http://www.springer.com/statistics/journal/13171

DOI: 10.1007/s13171-025-00406-9

Access Statistics for this article

Sankhya A: The Indian Journal of Statistics is currently edited by Dipak Dey

More articles in Sankhya A: The Indian Journal of Statistics from Springer, Indian Statistical Institute
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().