Model of hierarchical regression in decision problems of transportation systems

This article presents a method of multilevel hierarchical analysis for managing transportation systems – Intelligent Transport Systems (ITS). The task of multilevel planning it is essential to connect various links of the system. To make decisions a model of hierarchical structure has been applied as quite frequently actions in one point of the system affect results and actions taken in its other constituent. The article presents a method of estimation of the system parameters for the task of prediction. In the article the author is trying to prove that to select appropriate parameters to the model a weak correlation of independent features at simultaneous high correlation of these features with a dependent variable is necessary. All parameters necessary to solve decision problems must be random. A proposed model represents a compromise between estimation of the model in each group of routes separately and for all units of observations without taking into consideration grouping of routes