| Authors | Fatemeh Yousefzadeh,seyed ali asghar tajadod,Raynaldo Boris Arellano-Valle |
|---|---|
| Journal | Mathematics Interdisciplinary Research |
| Page number | 299-319 |
| Serial number | 11 |
| Volume number | 3 |
| Paper Type | Full Paper |
| Published At | 2026 |
| Journal Type | Typographic |
| Journal Country | Iran, Islamic Republic Of |
| Keywords | Kronecker, separable covariance, Multidimensional array, Tensor, EM algorithm, Image learning. |
Abstract
In recent years, rapid technological advances have led to the emergence of increasingly complex data structures, among which tensor data play a central role. To address the challenges of analyzing such data, it is essential to extend statistical concepts and methodologies to the tensor domain. In this paper, we introduce the tensor-variate t and skew-t distributions and derive several of their statistical properties. Parameter estimation is carried out using the Expectation–Maximization (EM) algorithm, and the performance of the proposed estimators is assessed through extensive simulation studies. Finally, the effectiveness and practical applicability of the estimators are demonstrated using real world data.