ARIMA模型与GM(1,1)模型在传染病发病率中的预测效果比较

Comparison of ARIMA model and GM(1,1)model in predicting the incidence of infectious diseases

为比较ARIMA模型和GM(1,1)模型对传染病的预测效果,对样本1(n=30)序列建立了GM(1,1)模型和ARIMA(p,d,q)模型,并根据平均相对误差(MRE)比较拟合模型的精度.得出结论:对于样本1,ARIMA(0,1,3)拟合度较好,MRE为4.68%;GM(1,1)模型MER值为15.00%,拟合效果较差.即ARIMA(0,1,3)模型拟合预测效果优于GM(1,1)模型,且预测结果表明未来病毒性肝炎发病率呈下降趋势.

In order to compare the predictive effects of the ARIMA model and the GM(1,1)model on infectious diseases,the GM(1,1)model and the ARIMA(p,d,q)model under sample 1(n=30)are established,and the average relative error(MRE)is used to compare the accuracy of the fitted model.It is concluded that for sample 1,ARIMA(0,1,3)has a good fit,and the MRE is 4.68%,the MER value of GM(1,1)model is 15.00%,and the fitting effect is poor.That is,the fitting prediction effect of the model ARIMA(0,1,3)is better than that of the GM(1,1)model,and the prediction results indicate that the incidence of viral hepatitis will decline in the future.