Citation

Park M. J. Transp. Res. (Seoul) 2017; 24(2): 27-39.

Copyright

(Copyright © 2017, Korea Transport Institute)

DOI

unavailable

PMID

unavailable

Abstract

The purpose of this study is to find the elements affecting traffic accident rates at three-way unsignalized intersections using a random parameters Tobit model. For accident rates, instead of using the number of accidents as a dependent variable, a Tobit regression model was applied. Through applying random parameters to enable the consideration of heterogeneity, specific characteristics of each intersection could also be analyzed. Modeling with random parameters was confirmed to have a slightly higher explanation power than modeling with fixed parameters, which is currently popularly performed. Nine variables were found to have statistically significant effects on accident rates and of them eight variables were found to have random parameters, indicating the effects on accident rates could be varied at each intersection. Therefore, it is recommended to consider the use of methods which could take heterogeneity into account concerning accident data in future traffic accident modeling research.

Language: ko

Keywords

Heterogeneity; Accident Rate; Random Parameter; Tobit Regression; Unsignalized Intersection