Streszczenie: W pracy przybliżono koncepcję modelu regresji Bayesa oraz przedstawiono wykorzystanie takiego modelu w budowaniu statystycznego klasyfikatora ciężkości wypadku drogowego w zależności od cech kierującego sprawcy. Modele Bayesa zostały wyznaczone dla dużej i małej próby treningowej z uwzględnieniem informatywnych i nieinformatywnych rozkładów a’priori parametrów strukturalnych oraz porównane z analogicznymi modelami klasycznymi MLE. Przedmiotowym klasyfikatorem statystycznym był model regresji logistycznej.
Abstract: The idea of a Bayes regression model was put forward and then the utilization of such a model while building a statistical classifier to identify a road accident severity in dependence on chosen at fault driver’s characteristics was presented in the paper. Bayes models were identified for small and big train samples assuming informative and non-informative prior distributions for structural parameters of the models. Obtained results were compared and referred to the results of classical MLE models. A logistic model was a statistical classifier under consideration.
B I B L I O G R A F I A1. El-Basyouny K., Barua S., Islam M. T. Investigation of time and weather effects on crash types using full Bayesian multivariate Poisson lognormal models. Accident Analysis and Prevention 201
73: 91-99.
2. Hand D., Manilla H., Padhraic S. Eksploracja danych. Warszawa: WN-T, 2005.
3. Häggström O. Finite Markov Chains and Algorithmic Applications. Cambridge University Press (Virtual Publishing) 2003.
4. Helai H., Chor C.H., Haque M.M. Severity of driver in jury and vehicle damage in traffic crashes at intersections: a Bayesian hierarchical analysis. Accident Analysis and Prevention 2008
40: 45-54.
5. Heydari S., Miranda-Moreno L.F., Lord D., Fu L. Bayesian methodology to estimate and update safety performance functions under limited data conditions: A sensitivity analysis. Accident Analysis and Prevention 2014
64: 41-51.
6. Huang H., Abdel-Aty M. Multilevel data and Bayesian analysis in traffic safety. Accident Analysis and Prevention 2010
42: 1556-1565.
7. Mitra S., Washington S. On the nature of over-dispersion in motor vehicle crash prediction models. Accident Analysis and Prevention 2007
39: 459-468.
8. Nowakowska M. Logistic models in the crash severity classification on the basis of chosen road characteristics. Transportation Research Record, Journal of the Transportation Research Board, Highway Safety Data, Analysis, and Evaluation, Volume 2, Washington D.C. 2010
2148: 16-26.
9. Nowakowska M. Modelowanie związków między cechami drogi a zagrożeniami w ruchu na drogach zamiejskich. Warszawa: Oficyna Wydawnicza Politechniki Warszawskiej, 2013.
10. Pei X. Wong S.C., Sze N.N. A joint probability approach to crash prediction models. Accident Analysis and Prevention 2011
43: 1160-1166.
11. Persaud B., Lan B., Lyon C., Bhim R. Comparison of empirical Bayes and full Bayes approach for before-after road safety evaluations. Accident Analysis and Prevention 2010
42: 38-43.
12. SAS/STAT® 9.2 User’s Guide (Introduction to Bayesian Analysis Procedures). Second Edition, Cary, NC, USA: SAS Institute Inc., 2009.
13. Yu R., Abdel-Aty M. Investigation different approaches to develop informative priors in hierarchical Bayesian safety performance functions. Accident Analysis and Prevention 2013
56: 51-58. 