Comparison of Three Prediction Models for the Incidence of Epidemic Diseases

Abstract

Nowadays, there are high incidences of epidemic diseases so it is very important to predict the incidence of them. There are many prediction methods for epidemic diseases at present. In various situations, different models have different applications. This article will select three prediction models, namely ARIMA model, grey model and BP neural network model. Taking the number of people infected by epidemics of Shandong from 2014 to 2019 as an example, based on the structure and performance of the model, it can be found that ARIMA model is suitable for the prediction of seasonal epidemics in schools and other densely populated places. The grey model needs less data and is suitable for the short-term prediction of some grass-roots prevention and control personnel. The BP neural network model has high prediction accuracy but complicated prediction process, and is suitable for the prediction of scientific research institutions.