时间序列分析(自回归求和移动平均模型)在流行性乙型脑炎预测中的应用

The Application of Autoregressive Integrated Moving Average Model on the Prediction of Japanese Encephalitis Cases

目的探讨时间序列分析[自回归求和移动平均(Autoregressive Integrated Moving Average,ARIMA)模型]在流行性乙型脑炎(乙脑)预测中的应用,建立乙脑报告发病数的预测模型,预测2010年乙脑发病趋势。方法使用2003年1月～2009年12月中国疾病监测信息报告系统中的乙脑报告月发病资料,使用SPSS软件专家建模器,考虑季节因素建立ARIMA预测模型,并用所得模型对2010年全国乙脑报告发病数进行预测。结果 ARIMA(0,0,1)(0,1,1)12模型能较好地拟合时间序列,2010年乙脑预测病例数为4579例,高峰仍在7、8月份。结论该ARIMA模型可较好地拟合乙脑发病的时间序列趋势;与2009年比较,预测2010年乙脑报告发病数相对平稳。

Objectives To explore the application of time series analysis[Autoregressive Integrated Moving Average（ARIMA） model]for Japanese Encephalitis（JE） cases prediction,and establish a predicting model for JE cases,predict the possible JE cases during the year 2010.Methods Reported JE cases by month from China Information System for Disease Control and Prevention since January 2003 to December 2009 were collected,established an ARIMA prediction model using the ＂Expert Modeler＂ in SPSS,considering the seasonal changes,then predict JE cases in year 2010.Results The model of ARIMA（0,0,1）（0,1,1）12 can appropriately fit the time series,4579 JE cases were predicted in year 2010,the peak is still at July and August.Conclusions The ARIMA model can be used to fit the time series trend of JE disease;Reported JE cases in year 2010 would be relatively stable,compared with the cases reported in year 2009.