Multivariate Non-Linear Time Series Modelling of Exposure and Risk in Road Safety Research

Abstract

A multivariate non-linear time series model for road safety data is presented. The model is applied in a case-study into the development of a yearly time series of numbers of fatal accidents (inside and outside urban areas) and numbers of kilometres driven by motor vehicles in the Netherlands between 1961 and 2000. The model accounts for missing entries in the disaggregated numbers of kilometres driven although the aggregated numbers are observed throughout. We consider a multivariate non-linear time series model for the analysis of these data. The model consists of dynamic unobserved factors for exposure and risk that are related in a non-linear way to the number of fatal accidents. The multivariate dimension of the model is due to its inclusion of multiple time series for inside and outside urban areas. Approximate maximum likelihood methods based on the extended Kalman filter are utilized for the estimation of unknown parameters. The latent factors are estimated by extended smoothing methods. It is concluded that the salient features of the observed time series are captured by the model in a satisfactory way. Copyright (c) 2010 Royal Statistical Society.

Road Safety Trends at National Level in Europe: A Review of Time-series Analysis Performed during the Period 2000–12

“…The analysis performed by the Dutch researchers was descriptive, and based on state space methods (Commandeur & Koopman, 2007). Their aim was to analyse changes in casualty data with focus on the following topics: taking into account the covariance structure in the analysis of casualty data sets of interdependent variables (Bijleveld, 2005); estimating risk, considered as a latent process, in a multivariate framework ; estimating both exposure and risk, considered as unobserved factors, and taking missing Road Safety Trends at National Level in Europe 659 exposure data into account (Bijleveld, Commandeur, Koopman, & van Montfort, 2010).…”

A Space–Time Multivariate Bayesian Model to Analyse Road Traffic Accidents by Severity

“…Examples of multivariate state space models in the area of road safety can be found in Bijleveld et al . () and Durbin and Koopman ().…”

Multivariate Forecasting of Road Traffic Flows in the Presence of Heteroscedasticity and Measurement Errors

“…The parameter μ t ( i ) is the unknown mean of Y t ( i ). When considering a similar variance modelling issue in DLMs in the related application of road safety research, Bijleveld et al. (2010) used the observations themselves as proxies for the unknown mean.…”

“…The parameter μ t .i/ is the unknown mean of Y t .i/. When considering a similar variance modelling issue in DLMs in the related application of road safety research, Bijleveld et al (2010) used the observations themselves as proxies for the unknown mean. In this paper, where the emphasis is very much on forecasting, μ t .i/ is estimated by its forecast, which is denoted f t .i/, obtained from the LMDM.…”