Analysis and forecast of the number of deaths, recovered cases, and confirmed cases from covid-19 for the top four affected countries using kalman filter

Abstract

COVID-19 is a virus that spread globally, causing severe health complications and substantial economic impact in various parts of the world. The COVID-19 forecast on infections is significant and crucial information that will help in executing policies and effectively reducing the daily cases. Filtering techniques are important ways to model dynamic structures because they provide good valuations over the recursive Bayesian updates. Kalman filters, one of the filtering techniques, are useful in the studying of contagious infections. Kalman filter algorithm performs an important role in the development of actual and comprehensive approaches to inhibit, learn, react, and reduce spreadable disorder outbreaks in people. The purpose of this paper is to forecast COVID-19 infections using the Kalman filter method. The Kalman filter (KF) was applied for the four most affected countries, namely the United States, India, Brazil, and Russia. Based on the results obtained, the KF method is capable of keeping track of the real COVID-19 data in nearly all scenarios. Kalman filters in the archetype background implement and produce decent COVID-19 predictions. The results of the KF method support the decision-making process for short-term strategies in handling the COVID-19 outbreak.