A nonlinear logit regression model for determining drivers’ accident probability in Algeria

The aim of the study is to estimate a nonlinear regression logit model to calculate the probability of a road accident occurring in Algeria. Referring to the literature, this probability depends on several variables related to the driver, the vehicle and the road environment. Thus, the objective is to show which explanatory variables favour the occurrence of a traffic accident and to what extent. In addition, we seek to estimate the probability that a driver with given characteristics, driving a vehicle with given characteristics, and having a driving licence for a given length of time may have an accident. The structure of the estimation is based on disaggregated data collected following the analysis of files proposed by an Algerian insurance company. The results obtained make it possible to determine the category of variables that has a significance and an important role in explaining the probability of accidents occurring.