Analysis of initial values of modeling parameters for distributed information systems

Optimization using the imitation models requires to analyse different procedures of statistical approximation. The immediate application of such models is extremely bulky and demands much machine time. The problems to simplify the simulation of the output imitation process, are considered. In order to simplify the simulation process, the classification of output imitation processes is performed according to the form of autocorrelation function and trend. The Gauss stationary process is proposed as a model of the output imitation process. It is shown that the proposed model covers a wide class of the real output imitation processes.