论文部分内容阅读
目的采用ARIMA模型预测手足口病的发病趋势,探讨提高模型预测准确性和实用性的思路。方法利用ARIMA模型对北京市西城区2010年1月—2015年12月手足口病月发病率进行拟合,并进行发病趋势预测。结果通过参数估计和残差白噪声检验,得到模型ARIMA(1,0,0)(1,1,0)_(12),BIC=1.958,Ljung—Box=7.885(P=0.952),2015年手足口病月发病率预测值和实际值的平均相对误差为21.47%。结论利用ARIMA模型预测手足口病的发病趋势是可行的,可以为地区传染病防控和公共卫生决策提供科学依据。
Objective To predict the incidence of HFMD using ARIMA model and to explore ways to improve the accuracy and practicability of the model. Methods ARIMA model was used to fit the monthly incidence rate of hand-foot-mouth disease in Xicheng District of Beijing from January 2010 to December 2015 and predict the incidence of HFMD. Results The parameters ARIMA (1,0,0) (1,1,0) _ (12), BIC = 1.958 and Ljung-Box = 7.885 (P = 0.952) were obtained through parameter estimation and residual white noise test. The average relative error between predicted and actual incidence of hand-foot-mouth disease was 21.47%. Conclusion Predicting the trend of HFMD using ARIMA model is feasible and can provide a scientific basis for prevention and control of public health and epidemic diseases in the region.