自回归滑动平均混合模型在辽宁省人间布鲁氏菌病发病预测中的应用  

Application of autoregressive integrated moving average model to predict human brucellosis incidence in Liaoning province

作　　者：王子江[1] 孙英伟[1] 毛玲玲[1] 于维君[1] 雷露[1] 孙广玖[1] 姚文清[2] Wang Zijiang;Sun Yingwei;Mao Lingling;Yu Weijun;Lei Lu;Sun Guangjiu;Yao Wenqing(Department of Infectious Disease Control and Prevention, Liaoning Provincial Center for Disease Control and Prevention, Shenyang 110005, Liaoning, China;Liaoning Provincial Center for Disease Control and Prevention, Shenyang 110005, Liaoning, China)

机构地区：[1]辽宁省疾病预防控制中心感染与传染性疾病防制所,辽宁沈阳110005 [2]辽宁省疾病预防控制中心,辽宁沈阳110005

出　　处：《疾病监测》2018年第3期198-202,共5页Disease Surveillance

基　　金：国家科技重大专项(No.2017ZX10103007)~~

摘　　要：目的了解辽宁省2006-2015年人间和家畜布鲁氏菌病(布病)疫情情况,建立时间序列模型,探索畜牧业因素对人间疫情影响,为布病预测防控提供依据。方法收集辽宁省2006-2016年布病月发病数,以发病数建立时间序列构建自回归滑动平均混合(ARIMA)模型,对2016年1-6月进行短期预测,结合实际观测值评价预测效果;收集辽宁省同期羊存栏量和疫情,分析与人间发病存在的关联。结果构建模型ARIMA(1,1,1)(1,1,1)12,月发病数拟合值与实际观测值时间序列基本一致,预测效果良好,但精确度较低。羊年存栏量与人间布病发病数存在正关联性。结论 ARIMA模型预测辽宁省布病发病情况基本可行,加强人畜布病的监测工作,以揭示人与家畜流行病学关系。ObjectiveTo establish an autoregressive integrated moving average（ARIMA） model to predict humanbrucellosis incidence in Liaoning,evaluate the influence of animal husbandry on human brucellosis and provide evidence for theprevention and control of human brucellosis.MethodsThe surveillance data of brucellosis from January 2006 to December2016 in Liaoning were collected for the establishment of ARIMA model to predict and validate the brucellosis incidence fromJanuary to June 2016 in Liaoning. The sheep stock data at the end of a year from 2006 to 2016 were collected to analysis itsrelationship with human brucellosis.ResultsThe ARIMA（1,1,1）（1,1,1）12 was established,and the predictive incidencewas similar to the actual incidence,but this model showed a lower accuracy. There was a positive correlation between sheepstock at the end of the year and the outbreak of human brucellosis.ConclusionIt was feasible to establish a ARIMA model forpredicting brucellosis incidence in Liaoning. It is important to strengthen human brucellosis surveillance and find out the theepidemiological correlation between human and livestock.

关 键 词：人间布鲁氏菌病 ARIMA模型 预测 

分 类 号：R516.7[医药卫生—内科学]