基于时间序列的交通事故死亡人数集成预测探究  

Integrated Prediction of Traffic Accident Deaths Based on Time Series

作　　者：李志成[1] 王珂 LI Zhi-cheng;WANG Ke(Department of Urban Rail Transit and Information Engineering, Anhui Communications Vocational &Technical College, Hefei Anhui 230001, China;School of Traffic and Transportation, Lanzhou Jiaotong University, Lanzhou Gansu 730070, China)

机构地区：[1]安徽交通职业技术学院城市轨道交通与信息工程系,安徽合肥230001 [2]兰州交通大学交通运输学院,甘肃兰州730070

出　　处：《菏泽学院学报》2022年第2期57-64,共8页Journal of Heze University

基　　金：2022年安徽省高校学科(专业)拔尖人才学术项目;安徽省高校自然科学研究重点项目(KJ2020A1064、KJ2021A1472);安徽省质量工程项目(2021kcszsfkc096,2020szsfkc030,2020zyq24)。

摘　　要：交通事故死亡人数预测对于我国交通发展具有重要意义.以我国道路交通安全状况为研究对象,探讨我国道路交通事故死亡人数及未来发展趋势.通过建立的VAR、VEC和ARMA模型,得出VAR模型和ARMA模型具有良好的预测效果;利用Johansen协整检验得出GDP、总人口数、公路里程、机动车保有量等影响因素与事故死亡人数存在长期均衡关系.最后提出基于VAR和ARMA模型的综合集成预测方法对我国2015~2020年的交通事故死亡人数进行集成预测,结果表明集成预测精度更高更稳定,能够有效进行交通事故死亡人数的短期预测.The prediction of traffic accident deaths is of great significance for China’s traffic development.This paper discusses the deaths and future development trend of road traffic accidents in China.The established VAR,VEC and ARMA models show that VAR model and ARMA model have good prediction effect.Johansen test shows that there is a long-term equilibrium relationship between GDP,total population,highway mileage,vehicle ownership and the number of accident deaths.The comprehensively integrated prediction method based on VAR and ARMA models is used to predict the traffic accident deaths in China from 2015 to 2020.The results show that the prediction accuracy is higher and more stable,and can make effective short-term prediction.

关 键 词：交通事故 VAR模型 VEC模型 ARMA模型 集成预测 

分 类 号：U491.3[交通运输工程—交通运输规划与管理]