作者: John Milton , Fred Mannering
关键词:
摘要: This research provides a demonstration of statistical model accident frequency that can eventually be used as part proactive program to allocate safety-related highway improvement funds. Negative binomial regressions annual on sections principal arterials in Washington State were estimated using data from two years (1992 and 1993). In all, 31306 observations estimation (annual frequencies specific highway). Our results isolated the effects various geometric traffic characteristics frequency. Subsequent elasticity computations identified relative importance variables included our specifications. The findings show negative regression this paper is powerful predictive tool one should increasingly applied future studies.