论文部分内容阅读
目的对河北省麻疹周报告疫情进行季节自回归滑动平均混合(SARIMA)模型分析,将应急预警时限提前,为更迅速地处置疫情提供依据。方法利用Eviews 8.0对河北省2004年第1周~2015年第24周麻疹报告发病数建立SARIMA模型,使用所建模型对历史疫情进行评估,对未来疫情进行预测预警。结果最终通过检验的最优模型是SARIMA(1,1,0)(2,1,2)_(52),表达式为(1+0.46B)(1+0.82B~(52)+0.16B~(104))d[ln(w_t+1),1,52]=(1+0.87B~(104))ε_t;Theil不等式系数=0.09,BP≈0,VP=0.01,CVP=0.99,模型拟合和预测良好。实际值均落在了预测值的95%可信范围内,对未来的预测与往年同期发病趋势一致。结论 SARIMA模型适用于河北省麻疹疫情的周报告疫情预警分析,能够捕捉到更为细微的疫情变化,预警结果更加精细。
Objective To analyze seasonal epidemiological report of measles in Hebei Province by seasonal autoregressive moving average mixed (SARIMA) model, advance the early warning time limit and provide basis for more rapid disposition of epidemic situation. Methods Eviews 8.0 was used to establish the SARIMA model for the incidence of measles reported from the first week of 2004 to the 24th of 2015 in Hebei Province from 2004 to 2004, and the historical epidemic situation was assessed using the model established to predict and predict the future epidemic. The optimal model for the final test results is SARIMA (1, 1, 0) (2,1,2) _ (52) and the expression is (1 + 0.46B) (1 + 0.82B ~ (52) + 0.16B (1 + 0.87B ~ (104)) ε_t; Theil inequality coefficient = 0.09, BP≈0, VP = 0.01, CVP = 0.99, and the model Fits and forecasts are good. The actual values all fall within the 95% confidence level of the predicted value, and the prediction of the future is consistent with the trend of the same period in previous years. Conclusion The SARIMA model is suitable for the early warning and analysis of the epidemic situation of measles in Hebei Province. It can capture the more subtle epidemic situation and make the warning result more detailed.