Discussion on Computationally Efficient Multivariate Spatio-Temporal Models for High-Dimensional Count-Valued Data by Bradley et al

TL;DR

This paper discusses a new multivariate spatio-temporal model for high-dimensional count data, introducing novel distributional results and connecting them with variational Bayes, while evaluating the model's performance through simulations.

Contribution

It presents a computationally efficient Poisson multivariate spatio-temporal mixed effects model with new distributional insights and links to variational inference methods.

Findings

The proposed P-MSTM performs well in simulations.

New distributional results on multivariate log-Gamma are derived.

Connections to mean field variational Bayes are established.

Abstract

I begin my discussion by summarizing the methodology proposed and new distributional results on multivariate log-Gamma derived in the paper. Then, I draw an interesting connection between their work with mean field variational Bayes. Lastly, I make some comments on the simulation results and the performance of the proposed Poisson multivariate spatio-temporal mixed effects model (P-MSTM).