COVID-19 Epidemic Forecasting

Abstract

Since the beginning of the COVID-19 pandemic, many dashboards have emerged as useful tools to monitor its evolution, inform the public, and assist governments in decision-making. We proposed a general methodology to produce forecasts on a one-week horizon, which is applicable to close to 200 countries, and as many states/regions or provinces. An additional challenge to achieve this goal is that the quality of the reported data varies significantly from country to country. This translates into different fluctuations and irregularities that can be observed in the reported time-series. Many countries do not report on a daily basis or delay their reports to particular days of the week. In particular, seasonal patterns with a weekly cycle are observed for many countries. It is important to note that (a), seasonal patterns are non-stationary and can actually change in time, in particular, if the reporting policies change. Furthermore, delays in reporting, changes in death cause attribution protocols, as well as changes in testing policies lead to abrupt corrections that introduce backlogs on some days, such that a number of daily cases or deaths which are anomalously high or even negative are reported. To take into account these peculiarities, we proposed a forecasting methodology that relies on estimating the underlying trend with a robust seasonal-trend decomposition method and using simple extrapolation techniques to make a forecast over a week.