论文部分内容阅读
目的探讨应用季节性ARIMA模型预测广州市手足口病发病情况的可行性。方法利用“传染病报告信息管理系统”数据,应用SPSS 13.0统计软件对广州市2009年1月至2014年6月手足口病发病数建立季节性ARIMA模型,使用所建模型对2014年7~12月发病情况进行预测。结果广州市手足口病发病特征以年为流行周期,每年有2个发病高峰;应用季节性ARIMA方法进行模型识别与估计后,建立ARIMA(1,0,1)(2,1,0)12模型,预测平均相对误差为0.22,预测效果较好。结论季节性ARIMA模型能较好的拟合广州市手足口病发病序列并进行短期的有效预测。
Objective To explore the feasibility of using seasonal ARIMA model to predict the incidence of HFMD in Guangzhou. Methods Using the “Infectious Diseases Reporting Information Management System” data, a seasonal ARIMA model was established using the SPSS 13.0 statistical software from Jan. 2009 to Jun. 2014 in Guangzhou City. Using the model, ~ December to predict the incidence. Results The incidence of hand, foot and mouth disease in Guangzhou was epidemic year with 2 peak incidences annually. After ARIMA method was used to identify and estimate the model, ARIMA (1,0,1) (2,1,0) 12 Model, the average relative error of prediction is 0.22, the prediction effect is better. Conclusion The seasonal ARIMA model can fit the incidence of hand-foot-mouth disease in Guangzhou and predict the short-term effective.