Statistical Analysis of Traffic Environment in the Udmurt Republic

Abstract

The paper covers the statistical analysis and prediction of the number of road accidents as well as the severity of their consequences in the Udmurt Republic. The cluster analysis of this region’s territories has been conducted and three groups of territories with different accident rates have been distinguished. The group with high accident rates included the territories with federal highways. The regression models of the dependence of the numbers of traffic accidents and people injured and killed during the accidents from qualitative and quantitative factors were built. The quantitative factors included the number of driving violations, the number of registered vehicles, the number of driver’s licenses issued for the first time, the number of cases of driving under the influence, the number of employees in the Directorate for Road Traffic Safety and the trend component. These models also take into account seasonal and territorial components which are qualitative factors included in the model after digitalization procedure of their quantitative version. To model the dynamics of exogenous quantitative factors the adequate autoregression models were chosen. Using the developed regression models and predictive dynamics of input variables the prediction of the number of traffic accidents and their consequences in the Udmurt Republic by the end of 2017 was made.