论文部分内容阅读
目的探讨应用求和自回归移动平均(Autoregressive Integrated Moving Average,ARIMA)模型进行手足口病发病数的拟合和预测,为手足口病疫情预警提供依据。方法运用统计分析软件Eviews 5.0对阜阳市2009年1月至2013年10月的手足口病发病的月发病数进行模型拟合,根据所建立的最优模型对2013年11月和12月的手足口病月发病数进行预测,并与实际值进行比较。结果阜阳市手足口病发病以年为周期,一年中4~7月份出现发病高峰。手足口病预测模型ARIMA(1,0,1)(0,1,1)12,检验结果总体最好,2013年11月至12月的手足口病月发病数的预测值与实际值较为接近,预测相对误差分别为3.35%和3.16%,预测效果较好。结论 ARIMA(1,0,1)(0,1,1)12模型可作为阜阳市手足口病月发病数中短期预测的模型。
Objective To investigate the fitting and prediction of the number of hand-foot-mouth disease using Autoregressive Integrated Moving Average (ARIMA) model, and to provide a basis for the early warning of hand-foot-mouth disease. Methods Using the statistical analysis software Eviews 5.0, the monthly incidence of HFMD in Fuyang from January 2009 to October 2013 was fitted with the model. According to the established optimal model, The incidence of stomatitis month was predicted and compared with the actual value. Results The incidence of hand, foot and mouth disease in Fuyang city was annual, and the incidence peak appeared from April to July in one year. ARIMA (1,0,1) (0,1,1) 12 is the best prediction model for hand-foot-mouth disease. The overall test results are the best. The predicted monthly incidence of hand-foot-mouth disease from November to December 2013 is close to the actual value , The relative error of prediction is respectively 3.35% and 3.16%, the prediction effect is better. Conclusion The ARIMA (1,0,1) (0,1,1) 12 model can be used as a model to predict the monthly incidence of HFMD in Fuyang City.