Task analysis identification of the distribution law of random processes

Abstract

Task analysis identification of the distribution law of random processes

Mikhaylov A. A., Bazuyeva S. A.

Incoming article date: 22.07.2015

In the article the general analysis identification of the problem for the law of the probability distribution of the observed random process (structural-parametric identification and comparative analysis of modern methods of identification of probability distributions showed, that at the present time to solve this problem is widespread method of normalized range (R/S analysis). Use this method when selecting a parametric model of the probability distribution best fits the experimental results, and minimal assumptions about the process under study and sufficient stability analysis of time series describing the investigated process. Information analysis method normalized range allowed us to propose a modified method for identification of the distribution law of random processes based on the normalized range (R/S analysis), for which the expressions for the elasticity and conditionality.

Keywords: diagnosis of technical systems, the method of normalized amplitude (RS analysis), method of Hirst, information analysis, the conditionality of the problem, the elasticity of the task