Abstract
The aim of the study is to apply mathematical methods to generate forecasts of the dynamics of random values of the percentage increase in the total number of infected people and the percentage increase in the total number of recovered and deceased patients. The obtained forecasts are used for retrospective forecasting of COVID-19 epidemic process dynamics in St. Petersburg and in Moscow. Materials and methods. When conducting a retrospective analysis and forecasting the dynamics of the total number of cases and the dynamics of the total number of patients who have either died or recovered, the values of percentage increases in these indicators were used. Retrospective analysis and forecasting of the dynamics of the COVID-19 epidemic process were carried out over 14-day time intervals, starting from March 25, 2020 to January 20, 2021, using the time series forecasting method proposed by the authors. Results and discussion. The retrospective two-week forecasts of the total number of cases and the number of active cases presented in the paper demonstrated a high accuracy performance, both in Moscow and St. Petersburg. The MAPE (mean absolute percentage error) for the total number of cases at the peaks of incidence, generally, did not exceed 1%. It is shown that the accuracy of the obtained retrospective forecasts of the total number of cases in St. Petersburg, built starting from May 2020, has increased significantly compared to the April forecasts. A similar conclusion can be made regarding the forecasts of the total number of cases in Moscow in April and May 2020.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call