Estimating a continuously varying offset between multivariate time series with application to COVID-19 in the United States

TL;DR

This paper presents new methods for continuously estimating the offset between multivariate time series and applies them to COVID-19 data in the US, revealing patterns that can inform healthcare resource planning.

Contribution

The paper introduces novel techniques for tracking time-varying offsets between multivariate time series, specifically applied to COVID-19 case and death data in the US.

Findings

Identified an 'up-down-up' pattern in the offset between cases and deaths.

Demonstrated potential for predicting healthcare system load.

Provided a framework for real-time offset estimation in epidemiology.

Abstract

This paper introduces new methods to track the offset between two multivariate time series on a continuous basis. We then apply this framework to COVID-19 counts on a state-by-state basis in the United States to determine the progression from cases to deaths as a function of time. Across multiple approaches, we reveal an "up-down-up" pattern in the estimated offset between reported cases and deaths as the pandemic progresses. This analysis could be used to predict imminent increased load on a healthcare system and aid the allocation of additional resources in advance.