ARIMA模型在浙江省2007—2017年其它感染性腹泻发病情况预测中的应用  

Application of ARIMA model in the prediction of incidence of other infectious diarrhea in Zhejiang province from 2007 to 2017

作　　者：高升辉 王明雯 李丹地[2] 李静欣[2] 李金松[2] 吴建军 段招军[2] Gao Shenghui;Wang Mingwen;Li Dandi;Li Jingxin;Li Jinsong;Wu Jianjun;Duan Zhaojun(School of Public Health,Gansu University of Chinese Medicine,Lanzhou 730099,China;National Institute for Viral Disease Control and Prevention,Chinese Center for Disease Control and Prevention,Beijing 102206,China)

机构地区：[1]甘肃中医药大学公共卫生学院,兰州730099 [2]中国疾病预防控制中心病毒病预防控制所,北京102206

出　　处：《国际病毒学杂志》2021年第5期384-387,共4页International Journal of Virology

基　　金：国家科技重大专项(2018ZX10201002)。

摘　　要：目的探讨应用差分整合移动平均自回归模型(autoregressive integrated moving average,ARIMA)分析预测2007—2017年浙江省其它感染性腹泻发病情况。方法利用国家人口与健康科学数据中心公共卫生科学数据中心提供的浙江省2007—2016年各月其它感染性腹泻发病人数的数据,用SPSS 25.0软件构建时间序列分析ARIMA模型,预测2017年每月的发病人数,并用该中心提供的实际值对模型进行评估。结果对基于2007—2016年其它感染性腹泻发病情况建立的ARIMA模型进行训练和序列验证,再通过建立Box-Ljung检验和BIC检验,最终确定ARIMA(0,1,2)(0,1,0)_(12)为非平稳时间序列最优模型,2017年的预测值与2017年实际数据对比,准确性较高。结论ARIMA(0,1,2)(0,1,0)_(12)模型对浙江省其它感染性腹泻流行的预测效果较好,预测结果将为其它感染性腹泻的监测和预防提供理论支撑。Objective To appraisal the application of autoregressive integrated moving average(ARIMA)model on prediction of the incidence of other infectious diarrhea in Zhejiang province from 2007 to 2017.Methods Based on monthly data of other infectious diarrhea cases in Zhejiang province from 2007 to 2016 provided by the Public Health Science Data Center of the National Population and Health Science Data Center,the ARIMA model was constructed by SPSS 25.0 statistic software to predict the monthly incidences of the cases in 2017.The model was evaluated by the differences between the predicted value and the actual value provided by the Center.Results The ARIMA model constructed from the data of other infectious diarrhea cases from 2007 to 2016 were trained and validated for the sequence.After the establishment of Box-Ljung test and BIC test,ARIMA(0,1,2)(0,1,0)_(12) was determined to be the best fitting model.The accuracy of predicted values in 2017 was higher compared to the actual data in 2017.Conclusions ARIMA(0,1,2)(0,1,0)_(12) model showed highly accurate prediction on the incidence of other infectious diarrhea cases in Zhejiang province.The results can be used as references for surveillance and early warning of other infectious diarrhea.

关 键 词：ARIMA 时间序列 其它感染性腹泻 发病数 预测 

分 类 号：R181.3[医药卫生—流行病学]