ARIMA模型在丙型肝炎月发病数预测中的应用

Application of ARIMA Model in Predicting the Monthly Incidence of Hepatitis C

目的探讨求和自回归移动平均模型(autoregressive integrated moving average,ARIMA)在丙肝发病趋势预测中的可行性,为卫生行政部门丙肝防控工作提供参考依据。方法运用SPSS 26.0软件对辽宁省朝阳市2009—2020年丙肝月发病数进行ARIMA模型的建模拟合,预测2021年1月至2021年12月朝阳市丙肝的月发病数,采用平均绝对百分比误差(MAPE)和均方根误差(RMSE)评价预测模型。结果根据2009—2019年每年朝阳市丙肝月发病数为基础建立时间序列,构建的最优模型为ARIMA(1,1,2)(1,1,2)_(12),该模型Ljung-BoxQ值为0.204,P>0.05,可认为残差序列为白噪声序列,模型各参数t检验P<0.05,参数检验具有统计学意义;运用该模型预测2018年、2020年每月的丙肝发病数,与实际发病数进行比较,得到MAPE分别为6.40%和9.41%,小于10%,RMSE分别为15.92和15.17,小于20,结果较小;预测2021年丙肝发病数为1885例。结论ARIMA模型能较好的预测朝阳市丙肝发病的变动趋势,2021年预测结果提示丙肝的发病呈上升趋势,为丙肝预防控制措施的制定提供一定的科学依据。

Objective To explore the feasibility of autoregressive integrated moving average(ARIMA)model in predicting the incidence trend of hepatitis C,and to provide reference basis for the prevention and control of hepatitis C in the health administration department.Methods SPSS 26.0 software was used to simulate the ARIMA model of the monthly incidence of hepatitis C in Chaoyang City,Liaoning Province from 2009 to 2020.The monthly incidence of hepatitis C in Chaoyang City from January to December 2021 was predicted.Mean absolute percentage error(MAPE)and root mean square error(RMSE)were used to evaluate the prediction models.Results The time series was established based on the number of monthly reported cases of tuberculosis in Chaoyang City from 2009 to 2019.The optimal model for ARIMA(1,1,2)(1,1,2)_(12) was constructed.The Ljung-BoxQ value of this model was 0.204(P>0.05),which could be considered as the residual sequence of white noise sequence.The t-test P value of each parameter of the model was all less than 0.05,which were statistically significant.The model was used to predict the number of hepatitis C cases per month in 2018 and 2020.Compared with the actual number of cases,a mean absolute percentage error was obtained,6.40%and 9.41%,less than 10%respectively;a root mean square error was obtained,15.92 and 15.17,less than 20,and the results were small.It was predicted that the number of hepatitis C cases in 2021 would be 1885.Conclusion ARIMA model can better predict the time series trend of hepatitis C incidence in Chaoyang City.The predicting outcomes of 2021 suggest that the incidence of hepatitis C is on the rise,which provides a scientific basis for the formulation of prevention and control measures for hepatitis C.