自回归求和移动平均模型季节乘积模型在青岛市肾综合征出血热发病预测中的应用

Application of multiple seasonal ARIMA model in forecasting the incidence of hemorrhagic fever with renal syndrome Qingdao city

目的探索ARIMA季节乘积模型在青岛市肾综合征出血热(HFRS)发病预测中的应用,为HFRS防治提供科学依据。方法利用1977~2015年青岛市HFRS月发病率,建立ARIMA季节乘积模型,以2016年月发病率评估预测效果,并预测2017年月发病率。结果青岛市HFRS发病具有明显的季节性,发病高峰为每年的10~12月。模型季节自回归参数为-0.45,BIC=3.58,平稳R2=0.98,残差序列检验为白噪声序列(q=13.56,P>0.05),建立ARIMA乘积季节模型ARIMA(0,3,2)(1,3,2)12,2016年实际值与拟合值绝对误差为0.01~0.28,且均在95%置信区间中,2017年月发病率0.15/10万~1.06/10万。结论 ARIMA乘积季节模型能够较好地模拟青岛市HFRS发病趋势,可用于短期预测该市HFRS发病情况。

Objective To explore the incidence of Hemorrhagic Fever with Renal Syndrome in Qingdao city by multiple seasonal ARIMA model of time series,so as to provide the basis for local prevention and control of HFRS. Methods A muhiple seasonal ARIMA model was fitted with data of monthly reported cases in Qingdao city, 1977-2015 for HFRS, evaluated by actual and predicted data in 2016,and forecasted data in 2017. Results The incidence of HFRS in Qingdao city was obviously seasonal, the peak of incidence was from October to December every year. Seasonal auto regressive coef- ficient was-0.45. BIC 3.58,steady R 2 was 0.98. Autocorrelation test for residuals of model was white-noise series（ P〉 0.05）. ARIMA （0,3,2）（1,3,2）12 was identified to fit and forecast monthly HFRS incidence. The absolute error between the actual value and the fitting value in 2016 was 0.01-0.28, and they were all in the 95% confidence interval. Monthly incidence rate of 2017 was 0.15/105-1. 06/105. Conclusion The multiple seasonal ARIMA model can be used to fit trends for incidence of HFRS in Qingdao city and forecast the incidence within a short period.