Full Text:   <2874>

Summary:  <1838>

CLC number: U491.31

On-line Access: 2014-07-08

Received: 2013-10-22

Revision Accepted: 2014-04-15

Crosschecked: 2014-06-24

Cited: 3

Clicked: 5365

Citations:  Bibtex RefMan EndNote GB/T7714

-   Go to

Article info.
Open peer comments

Journal of Zhejiang University SCIENCE A 2014 Vol.15 No.7 P.529-539

http://doi.org/10.1631/jzus.A1300342


Application of generalized estimating equations for crash frequency modeling with temporal correlation*


Author(s):  Wen-qing Wu1,2,3, Wei Wang1,2,3, Zhi-bin Li1,2,3, Pan Liu1,2,3, Yong Wang4

Affiliation(s):  1. Jiangsu Key Laboratory of Urban Intelligent Transportation Systems, Southeast University, Nanjing 210096, China; more

Corresponding email(s):   wangwei@seu.edu.cn

Key Words:  Crash frequency, Generalized estimating equation (GEE), Temporal correlation, Freeway, Safety


Wen-qing Wu, Wei Wang, Zhi-bin Li, Pan Liu, Yong Wang. Application of generalized estimating equations for crash frequency modeling with temporal correlation[J]. Journal of Zhejiang University Science A, 2014, 15(7): 529-539.

@article{title="Application of generalized estimating equations for crash frequency modeling with temporal correlation",
author="Wen-qing Wu, Wei Wang, Zhi-bin Li, Pan Liu, Yong Wang",
journal="Journal of Zhejiang University Science A",
volume="15",
number="7",
pages="529-539",
year="2014",
publisher="Zhejiang University Press & Springer",
doi="10.1631/jzus.A1300342"
}

%0 Journal Article
%T Application of generalized estimating equations for crash frequency modeling with temporal correlation
%A Wen-qing Wu
%A Wei Wang
%A Zhi-bin Li
%A Pan Liu
%A Yong Wang
%J Journal of Zhejiang University SCIENCE A
%V 15
%N 7
%P 529-539
%@ 1673-565X
%D 2014
%I Zhejiang University Press & Springer
%DOI 10.1631/jzus.A1300342

TY - JOUR
T1 - Application of generalized estimating equations for crash frequency modeling with temporal correlation
A1 - Wen-qing Wu
A1 - Wei Wang
A1 - Zhi-bin Li
A1 - Pan Liu
A1 - Yong Wang
J0 - Journal of Zhejiang University Science A
VL - 15
IS - 7
SP - 529
EP - 539
%@ 1673-565X
Y1 - 2014
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/jzus.A1300342


Abstract: 
Traditional crash frequency modeling uses crash frequency data averaged across multiple years. When data size is small, crash data in each year are used in the modeling to extend the size of the samples. The extension of sample size could create a temporal correlation among crash frequencies of the different years, which could affect the modeling accuracy. The primary objective of this study is to evaluate the application of the generalized estimating equation (GEE) procedures to account for the temporal correlation in the longitudinal crash frequency data. Four-year crash data at exit ramps on a freeway in China were collected for modeling. Based on the same data, traditional generalized linear models (GLMs) were estimated for model comparison. Results showed that traditional GLM underestimated the standard errors of coefficients for explanatory variables. The GEE procedure with an exchangeable correlation structure successively captured the temporal correlation among the crash frequencies of the different years. The GLM with GEE outperformed the traditional GLM in providing a good fit for the crash frequency data. Results of this study can help researchers better understand how various factors affect the crash frequencies at freeway divergent areas and propose effective countermeasures.

基于广义估计方程的时间相关性事故频次建模


研究目的:采用广义估计方程模型对存在时间相关性的事故频次数据进行建模,并与传统广义线性模型的估计效果进行对比。
创新要点:通过广义估计方程来考虑事故频次建模中数据的时间相关性,从而提高参数估计准确度以及模型预测精度。
研究方法:基于4年高速公路交通事故频次数据,建立考虑时间相关性的广义估计方程以及传统的广义线性模型,并采用统计指标对模型效果进行对比。
重要结论:1.事故频次数据样本量对预测精度影响很大;2.广义估计方程能够有效考虑事故频次数据中存在的时间相关性;3.广义估计方程的参数估计比传统广义线性模型更准确,且精度更高。

关键词:广义估计方程;事故频次;时间相关;广义线型模型

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

References

[1] Bared, J., Giering, G.L., Warren, D.L., 1999. Safety evaluation of acceleration and deceleration lane lengths. Institute of Transportation Engineers Journal, 69(5):50-54. 

[2] Bauer, K.M., Harwood, D.W., 1998.  Statistical Models of Accidents on Interchange Ramps and Speed-change Lanes. National Technical Information Service,Alexandria, USA :1-163. 

[3] Chen, H., Liu, P., Lu, J.J., 2009. Evaluating the safety impacts of the number and arrangement of lanes on freeway exit ramps. Accident Analysis and Prevention, 41(3):543-551. 


[4] Chen, H., Zhou, H., Zhao, J., 2011. Safety performance evaluation of left-side off-ramps at freeway diverge areas. Accident Analysis and Prevention, 43(3):605-612. 


[5] El-Basyouny, K., Sayed, T., 2009. Collision prediction models using multivariate Poisson-lognormal regression. Accident Analysis and Prevention, 41(4):820-828. 


[6] Hauer, E., 1997.  Observational Before-after Studies in Road Safety: Estimating the Effect of Highway and Traffic Engineering Measures on Road Safety. Pergamon Press, Elsevier Science Ltd.,Netherlands :

[7] Hauer, E., 2004. Statistical road safety modeling. Transportation Research Record: Journal of the Transportation Research Board, 1897(1):81-87. 


[8] Hu, G., Wen, M., Baker, T.D., 2008. Road-traffic death in China, 1985-2005: threat and opportunity. Injury Prevention, 14(2):129-130. 


[9] Jin, S., Huang, Z., Tao, P., 2011. Car-following theory of steady-state traffic flow using time-to-collision. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 12(8):645-654. 


[10] Jin, S., Wang, D., Ma, D., 2013. Using LBG quantization for particle-based collision detection algorithm. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 14(4):231-243. 


[11] Jones, A.P., Jrgensen, S.H., 2003. The use of multilevel models for the prediction of road accident outcomes. Accident Analysis and Prevention, 35(1):59-69. 


[12] Kim, D.G., Lee, Y., Washington, S., 2007. Modeling crash outcome probabilities at rural intersections: application of hierarchical binomial logistic models. Accident Analysis and Prevention, 39(1):125-134. 


[13] Li, Z., Liu, P., Wang, W., 2012. Using support vector machine models for crash injury severity analysis. Accident Analysis and Prevention, 45:478-486. 


[14] Li, Z., Chung, K., Cassidy, M., 2013. Collisions in freeway traffic: the influence of downstream queues and interim means to address it. Transportation Research Record: Journal of the Transportation Research Board, 2396(1):1-9. 


[15] Li, Z., Ahn, S., Chung, K., 2014. Surrogate safety measure for evaluating rear-end collision risk related to kinematic waves near freeway recurrent bottlenecks. Accident Analysis & Prevention, 64:52-61. 


[16] Liang, K.Y., Zeger, S.L., 1986. Longitudinal data analysis using generalized linear models. Biometrika, 73(1):13-22. 


[17] Liu, P., Chen, H., Lu, J., 2010. How arrangement of lanes on freeway mainlines and ramps affects safety of freeways with closely spaced entrance and exit ramps. Journal of Transportation Engineering, 136(7):614-622. 


[18] Lord, D., Persaud, B.N., 2000. Accident prediction models with and without trend: application of the generalized estimating equations procedure. Transportation Research Record: Journal of the Transportation Research Board, 1717(1):102-108. 


[19] Lord, D., Mannering, F., 2010. The statistical analysis of crash-frequency data: a review and assessment of methodological alternatives. Transportation Research Part A: Policy and Practice, 44(5):291-305. 


[20] Luoma, J., Sivak, M., 2012.  Road-safety Management in Brazil, Russia, India, and China. UMTRI-2012-1, University of Michigan Transportation Research Institute,Michigan, USA :

[21] Ma, J.M., Kockelman, K.M., 2006. Bayesian multivariate Poisson regression for models of injury count, by severity. Transportation Research Record: Journal of the Transportation Research Board, 1950(1):24-34. 


[22] Ma, S., Li, Q., Zhou, M., 2012. Road traffic injury in China: a review of national data sources. Traffic Injury Prevention, 13(sup1):57-63. 


[23] McCartt, A.T., Northrup, V.S., Retting, R.A., 2004. Types and characteristics of ramp-related motor vehicle crashes on urban interstate roadways in northern Virginia. Journal of Safety Research, 35(1):107-114. 


[24] Miaou, S.P., Lord, D., 2003. Modeling traffic crash-flow relationships for intersections: dispersion parameter, functional form, and Bayes versus empirical Bayes. Transportation Research Record: Journal of the Transportation Research Board, 1840(1):31-40. 


[25] Park, E., Lord, D., 2007. Multivariate Poisson-lognormal models for jointly modeling crash frequency by severity. Transportation Research Record: Journal of the Transportation Research Board, 2019(1):1-6. 


[26] Quddus, M.A., 2008. Time series count data models: an empirical application to traffic accidents. Accident Analysis and Prevention, 40(5):1732-1741. 


[27] Saenghaengtham, N., Kanongchaiyos, P., 2006. Using LBG quantization for particle-based collision detection algorithm. Journal of Zhejiang University-SCIENCE A, 7(7):1225-1232. 


[28] Shankar, V.N., Albin, R.B., Milton, J.C., 1998. Evaluating median cross-over likelihoods with clustered accident counts: an empirical inquiry using random effects negative binomial model. Transportation Research Record: Journal of the Transportation Research Board, 1635(1):44-48. 


[29] Wang, X., Abdel-Aty, M., 2006. Temporal and spatial analyses of rear-end crashes at signalized intersections. Accident Analysis and Prevention, 38(6):1137-1150. 


[30] Wang, X., Abdel-Aty, M., 2008. Modeling left-turn crash occurrence at signalized intersections by conflicting patterns. Accident Analysis and Prevention, 40(1):76-88. 


[31] Washington, S.P., Karlaftis, M.G., Mannering, F.L., 2010.  Statistical and Econometric Methods for Transportation Data Analysis. Chapman & Hall/CRC,Boca Raton, USA :


Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - 2024 Journal of Zhejiang University-SCIENCE