Traffic Safety Study: Empirical Bayes or Full Bayes?

In traffic safety studies, empirical Bayes (EB) methods have been widely applied for identifying hotspot locations and evaluating the effectiveness of countermeasures due to their sensible and uncomplicated construct. Recently, however, some traffic safety researchers have been advocating the full Bayes (FB) approach as a better alternative to EB for analyses of accident data. While it has been argued that FB is more flexible in accounting for full uncertainties in model parameters and accident risk estimates than EB, little is known about their relative merits and practical differences in terms of traffic safety analyses. This paper presents the results of a large scale simulation study on the performance of these two approaches as applied for hotspot identification. It was found that the full Bayes estimators performed better than the EB estimators when working with datasets with small number of sites (observations) and characterized by an overall low mean accident frequency. However, when the dataset is sufficiently large (e.g. over 300 sites), these two approaches yielded practically the same results.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; References; Tables;
  • Pagination: 25p
  • Monograph Title: TRB 86th Annual Meeting Compendium of Papers CD-ROM

Subject/Index Terms

Filing Info

  • Accession Number: 01044875
  • Record Type: Publication
  • Report/Paper Numbers: 07-1680
  • Files: BTRIS, TRIS, TRB
  • Created Date: Feb 8 2007 6:21PM