Impact of Heterogeneous Dispersion Parameter on the Expected Crash Frequency
作者： Kangwon Shin
DOI: 10.5762/KAIS.2014.15.9.5585
关键词:
摘要: This study tested the hypothesis that significance of heterogeneous dispersion parameter in safety performance function (SPF) used to estimate expected crashes is affected by endogenous prior distributions, and analyzed impacts mis-specified on evaluation results for traffic countermeasures. In particular, this simulated Poisson means based parameters estimated SPFs using both negative binomial (NB) model (HNB) analyzing mis-specification mean functions SPF. addition, characteristics errors crash reduction factors (CRFs) obtained when two models are posterior variances, which essentially through hyper-parameters distributions. The simulation showed a mis-estimation NB does not affect coefficient functions, but variances distribution seriously mis-estimated develop without considering heterogeneity dispersion. Consequently, erroneously distributions with parameters, can produce large CRFs up 120%.
koreascience.or.kr参考文章(13)
정은비, 오철, 속도를 이용한 고속도로 구간 사고율 예측 모형 Journal of the Eastern Asia Society for Transportation Studies. ,vol. 29, pp. 103- 111 ,(2011)
Kangwon Shin, Comparative Study on the Estimation Methods of Traffic Crashes: Empirical Bayes Estimate vs. Observed Crash Journal of the Korean Society of Civil Engineers D. ,vol. 30, pp. 453- 459 ,(2010)
E Haver, SELECTION FOR TREATMENT AS A SOURCE OF BIAS IN BEFORE-AND-AFTER STUDIES Traffic engineering and control. ,vol. 21, ,(1980)
Gang-Won Sin, Assessing Estimation Methods of the Expected Crashes using Panel Traffic Crash Data Journal of the Eastern Asia Society for Transportation Studies. ,vol. 29, pp. 103- 111 ,(2011)
Joseph M. Hilbe, Negative Binomial Regression ,(2007)
Srinivas Reddy Geedipally, Dominique Lord, Effects of Varying Dispersion Parameter of Poisson-Gamma Models on Estimation of Confidence Intervals of Crash Prediction Models Transportation Research Record. ,vol. 2061, pp. 46- 54 ,(2008) , 10.3141/2061-06
Dominique Lord, Peter Young-Jin Park, None, Investigating the effects of the fixed and varying dispersion parameters of Poisson-gamma models on empirical Bayes estimates. Accident Analysis & Prevention. ,vol. 40, pp. 1441- 1457 ,(2008) , 10.1016/J.AAP.2008.03.014
Tarek A Sayed, Karim El-Basyouny, Comparison of Two Negative Binomial Regression Techniques in Developing Accident Prediction Models Transportation Research Record. ,vol. 1950, pp. 9- 16 ,(2006) , 10.3141/1950-02
B.G. Heydecker, J. Wu, Identification of sites for road accident remedial work by Bayesian statistical methods: an example of uncertain inference Advances in Engineering Software. ,vol. 32, pp. 859- 869 ,(2001) , 10.1016/S0965-9978(01)00037-0