Evaluation and prediction of COVID-19 based on time-varying sir model

Abstract

This article based on the law of the COVID-19 epidemic situation until May 6 improves the SIR model. Through the reverse solution of the parameters in the model, the parameters are modeled and predicted, and the parameters that change with time are obtained. Compared with setting fixed parameters, the accuracy of the model is greatly improved. Using the improved model to analyze the COVID-19 epidemic situation and study the virus spreading trend in different countries. The results show that the improved model is basically reliable in predicting the development trend of the COVID-19 epidemic; the epidemic in Italy will basically end in July; the development trend in Britain and America is similar, and the inflection point is expected to appear in mid-June. The results of the study confirm the effectiveness of measures such as reducing the movement of people and providing medical assistance to the epidemic-stricken areas, and provide reference for subsequent epidemic prevention and control.