北京市昌平区肺结核发病数ARIMA模型预测  

Application of ARIMA model in predicting incidence of pulmonary tuberculosis in Changping district,Beijing

作　　者：原梅[1] 张治国[2] 豆智慧[3] 王路钦 张峣[3] 李卫民[4] 高基民[1] 

机构地区：[1]温州医科大学检验医学院生命科学学院浙江省模式生物技术与应用重点实验室,浙江温州325035 [2]北京市昌平区结核病防治所,北京102200 [3]中国疾病预防控制中心性病艾滋病预防控制中心,北京100050 [4]首都医科大学附属北京胸科医院北京市结核病胸部肿瘤研究所国家结核病临床实验室,北京101149

出　　处：《疾病监测》2015年第12期1045-1049,共5页Disease Surveillance

基　　金：国家自然科学基金(No.81273144);北京市卫生系统高层次卫生技术人才培养计划项目(No.20133041)~~

摘　　要：目的探讨自回归滑动平均混合模型(autoregressive integrated moving average model,ARIMA)模型在北京市昌平区肺结核发病数预测中的应用,阐述建模过程并预测2015年昌平区肺结核发病数,为制定防治策略合理配置资源等提供参考。方法采用全国结核病网络专报系统中2009 2014年现住址为北京市昌平区的肺结核报告发病数数据,通过模型识别、参数估计、检验诊断及模型评价,建立昌平区结核病发病数的ARIMA模型,并预测其2015年肺结核发病数。结果现住址为昌平区的肺结核发病数预测模型为ARIMA(0,1,1)(0,1,1)12,预测2015年的新发报告肺结核患者总数为851例,模型2015年第一、二季度(1 6月)预测误差率为1.65%,不到10%,模型预测精度较好。结论 ARIMA模型适用于昌平区肺结核发病数的早期预测。Objective To establish an ARIMA model for the prediction of incidence of pulmonary tuberculosis（ TB）in Changping district,Beijing in 2015 and provide evidence for the prevention and control of TB and resource allocation.Methods The incidence data of TB in local population in Changping during 2009- 2014 were collected from the National Tuberculosis Information Management System. An ARIMA model was established by means of model identification,parameter estimation,detection / diagnosis and model evaluation. The incidence of TB in Changping in2015 was predicted. Results With the established model of ARIMA（ 0,1,1）（ 0,1,1）12,it was predicted that the pulmonary TB case number would be 851 in local population in Changping in 2013. The error rate of the prediction during January-June 2015 was 1. 65%,lower than 10%,indicating the high precision of the model. Conclusion This study showed that the ARIMA model is applicable for the prediction of TB incidence in Changping.

关 键 词：ARIMA模型 结核病 发病数 预测 

分 类 号：R521[医药卫生—内科学]