"Back to the future" projections for COVID-19 surges

TL;DR

This paper introduces back-to-the-future (BTF) projections that leverage early surge data from other countries to accurately forecast future COVID-19 surges globally, outperforming traditional methods especially before inflection points.

Contribution

The paper presents a novel BTF projection method combining asynchronous time series matching with a response coaching SIR model to improve COVID-19 surge predictions.

Findings

BTF projections accurately predict surges across 12 countries.

Traditional methods often fail to predict future surges.

BTF cannot predict surges caused by new variants.

Abstract

We argue that information from countries who had earlier COVID-19 surges can be used to inform another country's current model, then generating what we call back-to-the-future (BTF) projections. We show that these projections can be used to accurately predict future COVID-19 surges prior to an inflection point of the daily infection curve. We show, across 12 different countries from all populated continents around the world, that our method can often predict future surges in scenarios where the traditional approaches would always predict no future surges. However, as expected, BTF projections cannot accurately predict a surge due to the emergence of a new variant. To generate BTF projections, we make use of a matching scheme for asynchronous time series combined with a response coaching SIR model.