skip to content

Faculty of Economics

Journal Cover

Harvey, A. C. and Kattuman, P.

A Farewell to R: Time Series Models for Tracking and Forecasting Epidemics

Covid Economics

(51) (2020)

Abstract: The time-dependent reproduction number, Rt, is a key metric used by epidemiologists to assess the current state of an outbreak of an infectious disease. This quantity is usually estimated using time series observations on new cases, or deaths, combining this information with the distribution of the serial interval of transmissions. For a new epidemic, such as COVID-19, the available information on the serial interval is limited. Bayesian methods are often used to combine this limited information with the new cases, with the new cases usually being smoothed by a simple, but to some extent arbitrary, moving average. This paper describes a new class of time series models for tracking and forecasting new cases. The viability of these models and their ability to deal with spikes and second waves is illustrated with data from Germany and Florida. As a by-product, estimates of Rt, together with their standard deviations, can be obtained from the growth rate of new cases. Very few assumptions are needed and those that are made can be checked. This leads us to the conclusion that tracking an epidemic by trying to estimate Rt may be neither necessary nor desirable.

Author links: Andrew Harvey  

Publisher's Link:

Papers and Publications

Recent Publications

Jochmans, K. and Weidner, M. Fixed-Effect Regressions on Network Data Econometrica [2019]

Galeotti, A., Golub, B. and Goyal, S. Targeting Interventions in Networks Econometrica [2020]

Agranov, M. and Elliott, M. Commitment and (in) Efficiency: A Bargaining Experiment Journal of the European Economic Association [2020]

Elliott, M., Georg, C-P. and Hazell, J. Systemic Risk Shifting in Financial Networks Journal of Economic Theory [2021]