The decisive problem in evaluating the passive vehicle safety from accident data by regression models is the assumption of linearity. Till now, all applied procedures combine the crash parameters linearly either to calculate the injury severity (linear regression) or to calculate the logit of the injury risk (logistic regression). These methods are not an appropriate tool for evaluating the passive vehicle safety since the influence of the crash parameters is obviously not linear. Most of the crash parameters exhibit nonlinear behaviour but it gets "linearized" by the mentioned linear regression models. Therefore, the influence of the crash variables is described wrongly and one gets falsified results for the evaluation of passive vehicle safety. The authors introduce a nonlinear nonparametric additive model to avoid the linearity of these models to evaluate the passive vehicle safety more appropriately. The additive model is applied to a database in order to examine the influence of different parameters on injury severity, and to demonstrate the nonlinear behaviour of parameters. In this application, only drivers of passenger cars which had a frontal collision are considered (300 cases). For the covering abstract of the conference see ITRD E203597.
[1]
Robert Zobel.
ACCIDENT DATA AND THE PASSIVE SAFETY OF VEHICLES OR CAN YOU RATE THE PASSIVE SAFETY OF VEHICLES FROM ACCIDENT DATA
,
1995
.
[2]
R. Tibshirani,et al.
Rejoinder: Linear Smoothers and Additive Models
,
1989
.
[3]
Stuart Vaughan Newstead,et al.
Measuring crashworthiness: make/model ratings and the influence of australian design rules for motor vehicle safety
,
1995
.
[4]
L Evans,et al.
Double pair comparison--a new method to determine how occupant characteristics affect fatality risk in traffic crashes.
,
1986,
Accident; analysis and prevention.
[5]
J. Friedman,et al.
Estimating Optimal Transformations for Multiple Regression and Correlation.
,
1985
.
[6]
Ivanka Horová,et al.
Linear Smoothers and Additive Models
,
1998
.