论文部分内容阅读
目的通过构建季节性时间序列模型(SARIMA),预测陕西省咸阳市日本脑炎疫情,为日本脑炎的预防控制提供参考。方法采用理论流行病学研究设计,收集2005年1月至2014年9月陕西省咸阳市日本脑炎的发病率资料,采用Box和Jenkins方法建立最佳的SARIMA模型,并对咸阳市日本脑炎的发病率进行短期预测。结果研究结果显示:SARIMA(1,1,1)(2,1,1)_(12)是最佳模型,该模型具有最低的贝叶斯信息标准(BIC),Akaike信息标准(AIC),平均绝对误差(MAE)值及最高的R~2和较低的平均绝对百分比误差(MAPE);SARIMA(1,1,1)(2,1,1)_(12)可对咸阳市日本脑炎发病率进行较为可靠的预测。结论 SARIMA(1,1,1)(2,1,1)_(12)可为咸阳市早期识别日本脑炎发病率异常升高(≥0.4/100000)提供预警;咸阳市日本脑炎的发病率将略有下降,从6月至8月达到顶峰。
Objective To predict the epidemic situation of Japanese encephalitis in Xianyang City, Shaanxi Province by constructing a seasonal time series model (SARIMA), and provide a reference for the prevention and control of Japanese encephalitis. Methods The theoretical epidemiological study design was used to collect the incidence data of Japanese encephalitis in Xianyang City of Shaanxi Province from January 2005 to September 2014. The best SARIMA model was established by Box and Jenkins method and the positive rate of Japanese encephalitis The incidence of short-term forecast. Results The results show that SARIMA (1,1,1) (2,1,1) _ (12) is the best model with the lowest BIC, Akaike information standard (AIC) Average absolute error (MAE) value and the highest R ~ 2 and lower mean absolute percentage error (MAPE); SARIMA (1,1,1) (2,1,1) _ (12) A more reliable prediction of the incidence of inflammation. Conclusions SARIMA (1,1,1) (2,1,1) _ (12) may provide early warning for the early identification of an abnormal increase of Japanese encephalitis (≥0.4 / 100000) in Xianyang City. The incidence of Japanese encephalitis in Xianyang City The rate will drop slightly, reaching its peak from June to August.