Please use this identifier to cite or link to this item:
|Title:||Risk assessment in traffic safety||Authors:||MOONS, Elke
|Issue Date:||2009||Source:||Joint Statistical Meetings 2009.||Abstract:||Traffic accidents are still a major cause of death all over the world. Within the age groups up to 44 years, it even holds an unfortunate third place. Hence, it is not surprisingly that traffic safety has become a 'hot' topic for policy makers, in the media, for academics as well as for a broad audience. However, in order to make the step from evidence to policy, risk models play an important role. This paper focuses on fitting increasingly complex models for the estimation of the relative accident risk in small areas, here applied to communities in Belgium. The most classical model is the Poisson model that will be extended to hierarchical models, such as the Poisson-Gamma or the Poisson-lognormal model, and even the spatial correlation structure will be taken into account via different ways. The results are compared to each other based on several goodness-of-fit measures.||Keywords:||spatial statistics; risk model; Bayesian mapping; Poisson regression; traffic safety;spatial statistics ; risk model ; Bayesian mapping ; Poisson regression ; traffic safety||Document URI:||http://hdl.handle.net/1942/10550||Link to publication:||http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=abstract_details&abstractid=304967||Category:||C2||Type:||Proceedings Paper|
|Appears in Collections:||Research publications|
Show full item record
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.