Scandinavian Working Papers in Economics
Memorandum,
Oslo University, Department of Economics
No 3/2023: Dynamic time series modelling and forecasting of COVID-19 in Norway
Gunnar Bårdsen () and Ragnar Nymoen
Additional contact information
Gunnar Bårdsen: NTNU
Ragnar Nymoen: Dept. of Economics, University of Oslo, Postal: Department of Economics, University of Oslo, P.O Box 1095 Blindern, N-0317 Oslo, Norway
Abstract: A framework for forecasting new COViD-19 cases jointly with hospital admissions and hospital beds with COVID-19 cases is presented. This project, dubbed CovidMod, produced 21-days ahead forecasts each working day from March 2021 to April 2022, and forecast errors that were used to assess forecast accuracy. A comparison with the forecasts of the Norwegian Institute of Public Health (NIPH), with dates of origin in the same period, favours the CovidMod forecasts in terms of lower RMSFEs (Root Mean Squared Forecast Errors), both for new cases and for hospital beds. Another comparison, with the short term forecasts (7 day horizon) produced by a forecasting project at the University of Oxford, shows only little difference in terms of the RMSFEs of new cases. Next, we present a further development of the model which allows the effects of policy responses to a central model parameter to be forecasted by an estimated smooth-transition function. The forecasting performance of the resulting non-linear model is demonstrated, and it is suggested as a possible way forward in the development of relevant forecasting tools in general and for pandemics in particular.
Language: English
23 pages, May 23, 2023
Full text files
memo-03-2023.pdfFull text
Questions (including download problems) about the papers in this series should be directed to Mari Strønstad Øverås ()
Report other problems with accessing this service to Sune Karlsson ().
This page generated on 2024-04-17 00:05:48.