基于分位结构模型的城市干道出入口管理安全影响因素识别
|
摘要
事故率及其影响因素的识别已有着多年的研究,但现有的方法大多采用均值回归模型来估计其参数。本文提出了采用分位结构模型作为分析事故率的方法,重点探讨城市主干道出入管理的异质性和内生性问题,用于识别基于出入管理技术的影响因素。采用两步估计方法,其中联立方程式模型针对事故率与行驶速度之间的内生性关系,而分位数回归模型不同于平均回归模型,可以预测不同事故率的分位数,用于解释未观测因素造成的异质性。与普通的联立方程模型相比,这种方法能够提供更全面的信息来阐明出入管理和影响因素对事故率的影响。本文利用美国内华达州交通运输局2013—2015年的事故数据来验证该模型的性能,并通过与一般联立方程组模型的比较,发现该模型不仅揭示了不同分位数的详细信息,也提高了预测精度。
The identification of crash rate and its influencing factors have been studied for years,but most of the existing methodologies adopt mean regression models to estimate the parameters. This study proposes the quantile structural equation model as a methodological alternative in analyzing crash rate,focusing on addressing the heterogeneity and endogeneity issues so as to identify the influencing factors of access management techniques in urban arterials. A two-step estimation procedure is performed,in which the simultaneous equations model accommodates the endogenous relationship between crash rate and travel speed,while the quantile regression estimates various quantiles of crash rate instead of the mean regression and accounts for the heterogeneity attributed to unobserved factors. The quantile approach is able to provide more comprehensive information about the impact of access management and other influencing factors on crash rate compared to the general simultaneous equations model. The crash dataset from 2013 to 2015 maintained by Nevada Department of Transportation is employed to illustrate the performance of proposed model. By comparing with the general simultaneous equations model,the proposed model reveals more detailed information in terms of different quantiles and improves the prediction accuracy.
The identification of crash rate and its influencing factors have been studied for years,but most of the existing methodologies adopt mean regression models to estimate the parameters. This study proposes the quantile structural equation model as a methodological alternative in analyzing crash rate,focusing on addressing the heterogeneity and endogeneity issues so as to identify the influencing factors of access management techniques in urban arterials. A two-step estimation procedure is performed,in which the simultaneous equations model accommodates the endogenous relationship between crash rate and travel speed,while the quantile regression estimates various quantiles of crash rate instead of the mean regression and accounts for the heterogeneity attributed to unobserved factors. The quantile approach is able to provide more comprehensive information about the impact of access management and other influencing factors on crash rate compared to the general simultaneous equations model. The crash dataset from 2013 to 2015 maintained by Nevada Department of Transportation is employed to illustrate the performance of proposed model. By comparing with the general simultaneous equations model,the proposed model reveals more detailed information in terms of different quantiles and improves the prediction accuracy.
引文
[1]美国交通运输研究委员会出入口管理分会.道路出入口管理手册[M].北京:中国建筑工业出版社,2009.
[2] Xu X,Kwigizile V,Teng H. Identifying access management factors associated with safety of urban arterials mid-blocks:a panel data simultaneous equation models approach[J]. Traffic Injury Prevention,2013, 14(7):734-742.
[3] Lord D,Mannering F. The statistical analysis of crashfrequency data:a review and assessment of methodological alternatives[J]. Transportation Research Part A:Policy and Practice,2010,44(5):291-305.
[4] Savolainen P T,Mannering F L,Lord D,et al. The statistical analysis of highway crash-injury severities:a review and assessment of methodological alternatives[J]. Accident Analysis&Prevention,2011,43(5):1666-1676.
[5] Mannering F L,Bhat C R. Analytic methods in accident research:methodological frontier and future directions[J]. Analytic Methods in Accident Research,2014,1:1-22.
[6]黄合来,许鹏鹏,马明,等.道路交通安全规划理论研究前沿[J].中国公路学报,2014,27(9):90-97.
[7] Wang Y,Feng X N,Song X Y. Bayesian quantile structural equation models[J]. Structructual Equation Modeling:A Multidisciplinary Journal,2016,23(2):246-258.
[8] Zhang Y,Tang N. Bayesian empirical likelihood estimation of quantile structural equation models[J]. Journal of Systems Science and Complexity, 2017, 30(1):122-138.
[9] Qin X,Ng M,Reyes P E. Identifying crash-prone locations with quantile regression[J]. Accident Analysis&Prevention,2010,42(6):1531-1537.
[10] Chernozhukov V,Fernández-Val I,Kowalski A E.Quantile regression with censoring and endogeneity[J]. Journal of Econometrics,2015,186(1):201-221.
[2] Xu X,Kwigizile V,Teng H. Identifying access management factors associated with safety of urban arterials mid-blocks:a panel data simultaneous equation models approach[J]. Traffic Injury Prevention,2013, 14(7):734-742.
[3] Lord D,Mannering F. The statistical analysis of crashfrequency data:a review and assessment of methodological alternatives[J]. Transportation Research Part A:Policy and Practice,2010,44(5):291-305.
[4] Savolainen P T,Mannering F L,Lord D,et al. The statistical analysis of highway crash-injury severities:a review and assessment of methodological alternatives[J]. Accident Analysis&Prevention,2011,43(5):1666-1676.
[5] Mannering F L,Bhat C R. Analytic methods in accident research:methodological frontier and future directions[J]. Analytic Methods in Accident Research,2014,1:1-22.
[6]黄合来,许鹏鹏,马明,等.道路交通安全规划理论研究前沿[J].中国公路学报,2014,27(9):90-97.
[7] Wang Y,Feng X N,Song X Y. Bayesian quantile structural equation models[J]. Structructual Equation Modeling:A Multidisciplinary Journal,2016,23(2):246-258.
[8] Zhang Y,Tang N. Bayesian empirical likelihood estimation of quantile structural equation models[J]. Journal of Systems Science and Complexity, 2017, 30(1):122-138.
[9] Qin X,Ng M,Reyes P E. Identifying crash-prone locations with quantile regression[J]. Accident Analysis&Prevention,2010,42(6):1531-1537.
[10] Chernozhukov V,Fernández-Val I,Kowalski A E.Quantile regression with censoring and endogeneity[J]. Journal of Econometrics,2015,186(1):201-221.