A Comparative Study on the Prediction Model of COVID-19

With the large-scale outbreak of the COVID-19 epidemic in 2020, scientists and medical workers around the world made contributions to reducing the epidemic outbreak, curing patients, and developing vaccines against the epidemic. In response to the COVID-19 epidemic in various countries, many scholars have used machine learning and infectious disease dynamics models to make corresponding predictions for the subsequent development of the epidemic. In order to acquire a more accurate forecast of the global epidemic, this paper used the ARIMA model, Logistic model, SIR model, and improved SEIR model to make predictions. Moreover, we compared these models’ benefits, drawbacks, and the degree of fit between the predicted data of various models and the real data to obtain a highly accurate model. Experiments have proved that among these models, the improved SEIR model can predict the future development trend of the global epidemic more precisely. In addition, the improved SEIR model also provides a certain theoretical basis for the government to issue a series of policies, prevention methods and control measures.