ARIMA模型在东莞市细菌性痢疾预测中的应用  

Application of autoregressive integrated moving average model in forecasting of bacillary dysentery in Dongguan

作　　者：张泽武[1] 曾耀明[1] 卢展鹏[1] 钟新光[1] 

机构地区：[1]东莞市疾病预防控制中心,广东东莞523129

出　　处：《中国热带医学》2013年第3期327-329,共3页China Tropical Medicine

摘　　要：目的探讨ARIMA模型预测东莞市细菌性痢疾发病的可行性和适用性,为东莞市细菌性痢疾的防控提供参考依据。方法使用SPSS17.0对2004年1月~2012年4月东莞市细菌性痢疾发病率资料拟合ARIMA模型,利用所得到的模型对东莞市2012年5月~7月细菌性痢疾发病率进行预测评价。结果 ARIMA（1,0,0）模型的预测值与实际值的平均相对误差为11.97%,实际值都在95%可信区间内,实际值与预测值变动趋势一致。结论 ARIMA（1,0,0）模型较好的反映了东莞市细菌性痢疾发病趋势,可作为东莞市细菌性痢疾发病水平短期预测模型。Objective To explore the feasibility of the application of autoregressive integrated moving average（ARIMA） model to forecast the incidence rate of bacillary dysentery in Dongguan city,and to provide basis for prevention and control.Methods SPSS13.0 was used to construct the ARIMA model based on the monthly incidence of bacillary dysentery in Dongguan city,from January 2005 to April 2012.Then the model was used to forecast the monthly incidence of dysentery from May to Junly in 2012,and the forecast result was compared with the actual incidence.Results The model of the monthly incidence rate of bacillary dysentery in Dongguan city was ARIMA（1,0,0）.The forecast monthly incidence rate of bacillary dysentery from May to Junly in 2012 consisted with the actual one,the relative error was11.97 %.Conclusion The ARIMA（1,0,0）model is appropriate in short-term forecasting of the incidence of bacillary dysentery in Dongguan city.

关 键 词：细菌性痢疾 自回归移动平均模型 预测 

分 类 号：R516.4[医药卫生—内科学]