为了从众多复杂的影响因素中单独剥离出智能交通建设项目对城市交通的改善作用，定量评估智能交通对城市交通运行效率提升的贡献，在充分借鉴政策公共效应双重差分评估模型的基础上，引入了匹配算法，创新性地提出了双重差分与倾向得分匹配相结合的定量评价模型。在此基础上，系统分析了影响城市交通运行效率的主要因素，并利用格兰杰因果检验，筛选出城市道路网平均运行车速、地区生产总值、民用汽车拥有量、道路长度、常住人口等作为模型评估的变量指标。然后选择36 个城市，采集了其2008—2012 年的基础数据进行对照。研究提出的基于匹配和双重差分的模型具有较强的适用性和稳健性，可用于量化评估智能交通建设对交通运输效率的改善作用。模型实证分析测算结果显示：智能交通建设在提升城市交通运行效率方面发挥了明显作用，对城市交通平均运行速度提升的贡献约在9.2%左右。
Abstract
In order to screen out the effect of ITS (Intelligent Transportation System) which is one of the various complex influencing factors on the improvement of urban traffic, and quantificationally evaluate the contribution of ITS to urban transport operating efficiency improvement, the matching algorithm was introduced into the traditional DID (Difference-in-difference) model which was usually used to evaluate the effect of the public policy. Then the quantitative evaluation model based on DID and PSM (Propensity Score Matching) was proposed to assess the contribution of ITS. The main factors affecting the efficiency of urban traffic were analyzed systematically, and the Granger Causality Test was used to select the most representative factors. Then the average running speed of urban road network, the GDP (Gross Domestic Product), the number of civil vehicles, the length of roads, and the resident population were chosen to be specified as the input parameters. The 2008-2012 basic data of 36 matching group cities were collected, then the simulated calculation was conducted. The Model basing on PSM and DID raised in this thesis has good generalizability and stability. Therefore it is helpful for evaluating the contribution of ITS projects to the transport efficiency improvement. The result of the simulation shows obvious improvement on urban transport operating efficiency from ITS, and it contributes about 9.2% to the improvement of the average speed of urban traffic.
关键词
智能交通; 交通效率 /
双重差分 /
倾向得分匹配 /
定量评价模型
Key words
ITS (Intelligent Transportation System) /
transport efficiency /
DID (Difference-in-difference) /
PSM (Propensity Score Matching) /
quantitative evaluation model
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 汪光焘. 城市交通与信息化[J]. 城市交通，2015（3）：1-4.
[2] 李腾海子. 交通信息化对北京交通运输业的贡献研究[D]. 北京：北京工业大学，2015.
[3] 董璐，云俊，李明伟. 基于DEA模型测算交通信息化对道路畅通性改善的贡献研究[J]. 数学的实践与认识，2016，46（16）：133-140.
[4] 吴利霞，王勇. 智能交通系统（ITS） 实施与城市交通畅通性改善——基于面板数据的双重差分估计分析[J]. 物流技术，2015，34（22）：113-116.
[5] 关淑琪，云俊，李明伟. ETC对武汉交通畅通性的贡献率研究[J]. 数学的实践与认识，2016，46（7）：81-86.
[6] HECKMAN J J, ICHIMURA H, TODD P E. Matching as an Econometric Evaluation Estimator: Evidence from Evaluating a Job Training Programme[J]. The Review of Economic Studies, 1997, 64(4): 605-654.
[7] HECKMAN J J, LOZANO S N. Using Matching, Instrumental Variables, and Control Functions to Estimate Economic Choice Models[J]. The Review of Economics and Statistics, 2004, 86(1): 30-57.
[8] CHAY K Y, MCEWAN P J, URQUIOLA M. The Central Role of Noise in Evaluating Interventions that Use Test Scores to Rank Schools[J]. The American Economic Review, 2005, 95(4): 1237-1258.
[9] 周黎安，陈烨. 中国农村税费改革的政策效果：基于双重差分模型的估计[J]. 经济研究，2005（8）：44-53.
[10] JALAN J, RAVALLION M. Estimating the Benefit Incidence of an Antipoverty Program by Propensity- Score Matching[J]. Journal of Business & Economic Statistics, 2003, 21(1): 19-30.
[11] 任其亮. 重庆市城市交通拥堵的原因及缓解对策[J]. 交通标准化，2009（13）：163-166.
[12] 石华军，楚尔鸣. 政策效果评估的双重差分方法[J]. 统计与决策，2017（17）：80-83.
[13] 马雄威. 线性回归方程中多重共线性诊断方法及其实证分析[J]. 华中农业大学学报（社会科学版），2008（2）：78-81，85.
[14] BROOKHART M A, SCHNEEWEISS S, ROTHMAN K J, et al. Variable Selection for Propensity Score Models[J]. American Journal of Epidemiology, 2006, 163(12): 1149-1156.
[15] 黄玲文，姚洋. 国有企业改制对就业的影响——来自11个城市的证据[J]. 经济研究，2007（3）：57-69.
[16] 胡永远，周志凤. 基于倾向得分匹配法的政策参与效应评估[J]. 中国行政管理，2014（1）：98-102.
[17] 王长青，张一农，许万里. 运用最小二乘法确定后评估指标权重的方法[J]. 吉林大学学报（信息科学版），2010，28（5）：513-518.
京公网安备 11010102003417号
京ICP备17006742号-3
