Observatorio de I+D+i UPM

Memorias de investigación
Communications at congresses:
An application of dynamic statistical models to van-involved accidents in Spain: a comparison with different alternatives within the UCM and DRAG frameworks
Research Areas
  • Engineering
Aggregated econometric models are a very useful tool to analyze the evolution of road accidents and the effect of exposure, socioeconomic factors and measures taken by the administration on accident rates and severities. There are different sources of uncertainty which require the application of a statistical approach, where time-series models derived from the original Box-Jenkins framework are preferably selected. The models present different degrees of sophistication, and the simplest are chosen whenever possible, thus facilitating interpretation and respecting the principle of parsimony. Here we focus on two very relevant model categories: the Demand for Road use, Accidents and their Gravity (DRAG) methodology, developed by Gaudry (1984) and Unobserved Components Model (UCM) with intervention proposed by Harvey and Durbin (1986) were some effects are modeled through latent variables within a state-space framework. Here we apply the two models to the general Spanish Accident Data Base whenever vans are involved and compare the results in terms of goodness of fit, in the framework of the research Project FURGOSEG - P24/08 ?Development and application of an integrated methodology for the study of the traffic accidents with LTVs involved?, of the Spanish National Research Plan 2008-2011.
Congreso Ingeniería del Transporte CIT 2012
Start Date
End Date
From page
To page
Libro de actas
  • Autor: Bahar Dadashova . (UPM)
  • Autor: Blanca del Valle Arenas Ramirez (UPM)
  • Autor: Jose Manuel Mira Mcwilliams (UPM)
  • Autor: Francisco Aparicio Izquierdo (UPM)
Research Group, Departaments and Institutes related
  • Creador: Grupo de Investigación: Grupo de Inv. en Seguridad e Impacto Medioambiental de Vehículos y Transportes (GIVET)
  • Grupo de Investigación: Estadística computacional y Modelado estocástico
  • Centro o Instituto I+D+i: Instituto Universitario de Investigación del Automóvil (INSIA)
  • Departamento: Ingeniería Mecánica y de Fabricación
  • Departamento: Ingeniería de Organización, Administración de Empresas y Estadística
S2i 2019 Observatorio de investigación @ UPM con la colaboración del Consejo Social UPM
Cofinanciación del MINECO en el marco del Programa INNCIDE 2011 (OTR-2011-0236)
Cofinanciación del MINECO en el marco del Programa INNPACTO (IPT-020000-2010-22)