摘要
目的探讨应用自回归求和滑动平均(autoregressive integrated moving average,ARIMA)模型预测浦东新区手足口病发病率的可行性。方法基于2010—2015年浦东新区的逐月手足口病发病率,建立最优ARIMA模型,回代预测2015年的手足口病发病率,比较预测值与实际值的差异,并预测2016年的手足口病发病率。结果模型ARIMA(1,1,1)(1,1,0)_(12)较好拟合了既往手足口病发病率的时间序列,2015年逐月发病率的预测值符合实际值的变动趋势,全年发病率预测值与实际值的相对误差率为1.81%。预测2016年浦东新区手足口病发病率为269.97/10万。结论 ARIMA模型可用于短期预测未来的手足口病发病率。
Objective To explore the feasibility for the application of time series ARIMA model to predict the incidence rate of handfoot-and-mouth（ HFM） disease in Pudong New District,Shanghai. Methods ARIMA model was established based on the monthly incidence rate of HFM disease of Pudong New District from 2010 to 2015. The constructed optimal model was applied to predict the incidence rate of HFM disease of Pudong in 2015,and evaluate the validity of model through comparing the difference of predicted incidence rate and actual one. The incidence rate of HFM disease in 2016 was predicted by ARIMA model based on the incidence rate from Jan 2010 to Dec 2015. Results The model ARIMA（ 1,1,1）（ 1,1,0）_（12）could preferably fit the time series of incidence rate. The predicted incidence rate in 2015 was consistent with the actual one with the relative error of 1. 81%. The predicted incidence rate of HFM disease in 2016 based on the incidence rate from 2010 to 2015 would be 269. 97 per100 thousand. Conclusion ARIMA model is a predicted model of high precision for short-time forecast.
出处
《公共卫生与预防医学》
2016年第3期32-36,共5页
Journal of Public Health and Preventive Medicine
