ABSTRACT
The Empirical Bayes (EB) approach to road safety estimation has developed gradually during the past 30 years. It is now firmly established and recommended as the state-of-the-art approach to the estimation of the expected number of collisions. The EB approach was originally developed for the purpose of controlling for the “regression-to-mean” problem in before-and-after studies evaluating the effects of road safety measures. This has remained an important area of application, but EB methods are now also used to identify hazardous road locations. This chapter outlines several versions of EB that exist now. Developing good collision prediction models is a complex process, and there are many analytic choices. A case study on the usage of motorcycle helmets in a Chinese city will be included to illustrate the methods.
