Exploration of fully Bayesian methods for observational before-after road safety studies
The objective of this study was to explore the application of fully Bayesian methods for before-after road safety studies. The regular before-after fully Bayesian method, which uses only reference sites to develop models, and the intervention before-after fully Bayesian method, employing both reference sites and treated sites to develop models, are presented in this paper. The methods were evaluated with a simulated dataset through randomly assigning hypothetical treatments to high accident sites to mimic a common site selection process in road jurisdictions. It was confirmed that both methods can provide promising results in accounting for the regression-to-the-mean that results from this biased site selection process.
-
Authors:
- Lan, B
- Persaud, B
- Conference:
- Publication Date: 2007-6
Language
- English
Media Info
- Pagination: 15p
- Monograph Title: Canadian Multidisciplinary Road Safety Conference XVII, June 3-6, 2007, Montreal, Quebec
Subject/Index Terms
- TRT Terms: Before and after studies; High risk locations; Highway safety; Statistical analysis; Traffic engineering
- Uncontrolled Terms: Road safety (engineering and vehicles)
- ATRI Terms: Before and after study; Crash black spot; Road safety; Statistical analysis; Traffic engineering
- Subject Areas: Data and Information Technology;
Filing Info
- Accession Number: 01386950
- Record Type: Publication
- Source Agency: ARRB
- Files: ATRI
- Created Date: Aug 22 2012 9:56PM