Missing observation analysis for matrix-variate time series data

TL;DR

This paper develops Bayesian inference methods for matrix-variate dynamic linear models to effectively handle missing observations in any sub-matrix of the data, through distribution modifications and algorithm updates.

Contribution

It introduces new distribution modifications and an updated algorithm enabling missing data analysis in matrix-variate time series models.

Findings

Effective handling of missing sub-matrix observations.

Modified distributions facilitate Bayesian inference.

Updated algorithms improve missing data analysis.

Abstract

Bayesian inference is developed for matrix-variate dynamic linear models (MV-DLMs), in order to allow missing observation analysis, of any sub-vector or sub-matrix of the observation time series matrix. We propose modifications of the inverted Wishart and matrix $t$ distributions, replacing the scalar degrees of freedom by a diagonal matrix of degrees of freedom. The MV-DLM is then re-defined and modifications of the updating algorithm for missing observations are suggested.