ARIMA模型在涂阳肺结核月发病预测中的应用  

Application of ARIMA model in forecasting monthly incidence of smear-positive tuberculosis

作　　者：杨海琴[1] 胡代玉 刘瑛 王润华[1] 易静[1] 

机构地区：[1]重庆医科大学公共卫生与管理学院卫生统计与信息管理学教研室,重庆400016 [2]重庆市结核病防治所,重庆400020

出　　处：《第二军医大学学报》2013年第9期980-984,共5页Academic Journal of Second Military Medical University

基　　金：国家自然科学基金(30872160);重庆市科委自然科学基金(CSTC;2009BB5415)~~

摘　　要：目的探讨重庆市涂阳肺结核月发病数随时间的变化规律,为控制和预防肺结核提供科学依据。方法采用SPSS13.0软件对2005～2009年重庆市涂阳肺结核月发病数资料建立ARIMA模型,利用该模型预测2010年1月～12月的涂阳肺结核月发病数,对模型的短期预测及其效果进行初步评价。结果建立的ARIMA(1,1,0)×(0,1,1)12模型是拟合重庆市涂阳肺结核月发病数的合适模型,2005～2009年观测值落在拟合值95%的可信区间内,2010年预测值的平均相对误差为6.31%。结论 ARIMA(1,1,0)×(0,1,1)12模型能很好地预测重庆市涂阳肺结核月发病情况,为控制和预防肺结核提供了可靠依据。Objective To investigate the variation of the monthly incidence of smeavpositive tuberculosis with time in Chongqing, so to provide a scientific evidence for the control and prevention of tuberculosis. Methods Using the SPSS 13.0 software, we established an ARIMA model with the monthly incidence data of smear-positive tuberculosis （2005-2009）, and the model was used to forecast the monthly incidence of Jan. 2010 to Dec. 2010. The short-term forecasting efficacy was evaluated. Results The established ARIMA （1,1,0）× （0,1,1）12 model was suitable for forecasting the monthly incidence of smear- positive tuberculosis in Chongqing. The observed values of 2005 2009 were in the 95 % confidence interval of the fitted values, and the average relative error of the predictive value was 6.31% for 2010. Conclusion ARIMA （1,1,0） × （0,1,1）12 model can satisfactorily forecast the monthly incidence of smear-positive tuberculosis in Chongqing, which provides a reliable evidence for control and prevention of tuberculosis.

关 键 词：ARIMA模型 肺结核 预测 发病率 

分 类 号：R521[医药卫生—内科学]