Skip to main content
Article
Classification with the matrix-variate-t distribution
arXiv
  • Geoffrey Z. Thompson, Iowa State University
  • Ranjan Maitra, Iowa State University
  • William Q. Meeker, Iowa State University
  • Ashraf Bastawros, Iowa State University
Document Type
Article
Publication Version
Submitted Manuscript
Publication Date
1-1-2019
Abstract

Matrix-variate distributions can intuitively model the dependence structure of matrix-valued observations that arise in applications with multivariate time series, spatio-temporal or repeated measures. This paper develops an Expectation-Maximization algorithm for discriminant analysis and classification with matrix-variate t-distributions. The methodology shows promise on simulated datasets or when applied to the forensic matching of fractured surfaces or the classification of functional Magnetic Resonance, satellite or hand gestures images.

Comments

This is a pre-print of the article Thompson, Geoffrey Z., Ranjan Maitra, William Q. Meeker, and Ashraf Bastawros. "Classification with the matrix-variate-t distribution." arXiv preprint arXiv:1907.09565 (2019). Posted with permission.

Copyright Owner
The Authors
Language
en
File Format
application/pdf
Citation Information
Geoffrey Z. Thompson, Ranjan Maitra, William Q. Meeker and Ashraf Bastawros. "Classification with the matrix-variate-t distribution" arXiv (2019)
Available at: http://works.bepress.com/ashraf-bastawros/33/