灰色GM(1,1)模型在结核病发病率预测中应用  

Application of gray GM (1,1) model to predict incidence of tuberculosis in Henan province

作　　者：李立[1,2] 杨召[1] 叶中辉[1] 郭奕瑞[1] 梁淑英 尤爱国[3] 张肖肖[3] 王重建[1] 

机构地区：[1]郑州大学公共卫生学院流行病与卫生统计学系,河南郑州450001 [2]河南省医学科学院 [3]河南省疾病预防控制中心

出　　处：《中国公共卫生》2014年第4期396-397,共2页Chinese Journal of Public Health

基　　金：国家自然科学基金(81001293);河南省科技攻关计划项目(122102310210)

摘　　要：目的应用灰色GM(1,1)模型拟合结核病发病率,探讨其在结核病发病率预测中的应用。方法利用河南省2004—2011年结核病疫情资料,建立结核病发病率灰色GM(1,1)预测模型,评价模型预测效能,预测该省2012—2014年结核病发病率。结果河南省结核病发病率建立的灰色GM(1,1)预测模型的平均相对误差、后验差比值(C)、小误差概率(P)及平均级比偏差值(珔P)分别为3.71%、0.21、1.00和0.0269,模型预测效能较好,利用该模型对2012—2014年结核病发病率进行外推预测,结果分别为65.82/10万、56.42/10万、47.31/10万。结论灰色GM(1,1)模型对于结核病发病率的预测效能较好,预测结果对于结核病预防控制具有重要指导意义。Objective To develop a gray GM （1,1）model and to explore its potential application in prediction of tuberculosis incidence.Methods A gray GM （1,1）model was established using the epidemic data of tuberculosis in Henan province from January 1,2004 to December 31,2011,and the predictive performance was tested and accessed.Results The average relative error,posterior margin ratio,small error probability,and average level deviation were 3.71%,0.21,1.00,and 0.0269,respectively,suggesting the gray GM （1,1）model could be applied for predicting tuberculosis incidence.Based on the model,the tuberculosis incidence predicted for the province form 2012 to 2014 were 65.82/105,56.42/105,and 47.31/105,respectively.Conclusion The gray GM （1,1）model could be used to predict the incidence of tuberculosis for the prevention management and measurement of the disease.

关 键 词：结核病 发病率 灰色GM(1 1)模型 预测 

分 类 号：R52[医药卫生—内科学]