Abstract:
On March 11, 2020, the WHO declared that the SARS-CoV-2 virus causing COVID-19 had become a pandemic. Since then, infections have overwhelmed many health systems, and sadly in some countries, also mortuaries. As accurate forecasts could help health and public authorities plan better, in mid-March 2020, the authors started to produce short-term forecasts for the number of recorded cases and deaths from COVID-19, producing about four updates a week on www.doornik.com/COVID-19. We now have experience in producing regular forecasts for over a year, so in this article we discuss the main measurement and methodological challenges faced when modeling and forecasting COVID-19 data.
Citation:
Doornik, J.A., Castle, J.L., & Hendry, D.F. (2021) "Modelling and forecasting the COVID-19 pandemic time-series data", Social Science Quarterly, Special Issue: Public Policy, Behavior, and Health Outcomes during the COVID Pandemic. https://doi.org/10.1111/ssqu.13008