EM Estimation of Conditional Matrix Variate $t$ Distributions

TL;DR

This paper introduces a new version of the conditional matrix variate Student $t$ distribution and develops EM algorithms for parameter estimation, including special cases with Minnesota prior, advancing statistical modeling techniques.

Contribution

The paper proposes a novel version of the conditional matrix variate Student $t$ distribution and provides EM algorithms for efficient parameter estimation, including special cases with prior.

Findings

Developed EM algorithms for the new distribution

Extended methods to include Minnesota prior cases

Enhanced statistical modeling capabilities

Abstract

Conditional matrix variate student $t$ distribution was introduced by Battulga (2024a). In this paper, we propose a new version of the conditional matrix variate student $t$ distribution. The paper provides EM algorithms, which estimate parameters of the conditional matrix variate student $t$ distributions, including general cases and special cases with Minnesota prior.