Predicting COVID-19 hospitalisation using a mixture of Bayesian predictive syntheses

TL;DR

This paper introduces the mixture of Bayesian predictive syntheses (MBPS), a new method for predicting COVID-19 hospitalizations at the subnational level by clustering time series data and combining predictive models.

Contribution

The paper presents MBPS, a novel clustering-based Bayesian approach that improves COVID-19 hospitalization predictions without requiring complex multivariate count models.

Findings

MBPS improves predictive accuracy over existing methods.

MBPS provides better uncertainty quantification.

Effective in modeling COVID-19 hospitalization data in Japan and Korea.

Abstract

This paper proposes a novel methodology called the mixture of Bayesian predictive syntheses (MBPS) for multiple time series count data for the challenging task of predicting the numbers of COVID-19 inpatients and isolated cases in Japan and Korea at the subnational-level. MBPS combines a set of predictive models and partitions the multiple time series into clusters based on their contribution to predicting the outcome. In this way, MBPS leverages the shared information within each cluster and is suitable for predicting COVID-19 inpatients since the data exhibit similar dynamics over multiple areas. Also, MBPS avoids using a multivariate count model, which is generally cumbersome to develop and implement. Our Japanese and Korean data analyses demonstrate that the proposed MBPS methodology has improved predictive accuracy and uncertainty quantification.