我国交通事故死亡人数的组合测度及关联性分析  

The combined measurement and correlation analysis of deaths in traffic accidents

作　　者：孙浩[1] 杨桂元[1] SUN Hao;YANG Guiyuan(School of Statistics and Applied Maths,Anhui University of Finance and Economic,Bengbu 233030,China)

机构地区：[1]安徽财经大学统计与应用数学学院,安徽蚌埠233030

出　　处：《延边大学学报（自然科学版）》2018年第3期239-245,共7页Journal of Yanbian University（Natural Science Edition）

基　　金：国家社会科学基金资助项目(12BTJ008)

摘　　要：为了提高对交通事故的预测精度,在4种单项方法(多元线性回归、Brown指数平滑法、Holt指数平滑法、GM(1,n))的基础上,构建了基于Theil不等系数的IOWHA算子的最优组合预测模型.对我国1997—2016年的交通事故死亡人数进行预测的结果表明,本文提出的组合预测模型优于各单项方法,是一种优性组合预测模型.采用VAR模型分别对等级公路里程、国民总收入、机动车驾驶员人数与交通事故死亡人数间的关系进行脉冲响应研究表明,国民总收入对交通事故死亡人数具有抑制作用,而等级公路里程与机动车驾驶员人数的增加则会增加交通事故的死亡率.In order to improve the prediction accuracy of traffic accidents,an optimal combination prediction model based on the IOWHA operator with unequal coefficients of Theil was constructed on the basis of four single methods(multiple linear regression,Brown index smoothing,Holt index smoothing,GM(1,n)).The prediction results of traffic accident deaths in China from 1997 to 2016 show that the combined prediction model proposed in this paper is superior to each single method and is a superior combined prediction model.The relationship between grade highway mileage,gross national income,the number of motor vehicle drivers and the number of traffic deaths was studied by using VAR model,the results show that the gross national income has inhibitory effect on deaths,and grade highway mileage and the increase of motor vehicle driver traffic can increase mortality.

关 键 词：交通事故 Theil不等系数 IOWHA算子 VAR模型 

分 类 号：F224.0[经济管理—国民经济]