Methods of statistical data processing in problems of identification of dynamic systems

In this article the problem of identifying dynamic systems is considered. Here is provided infographic which describing the numbers of scientific works on a given topic. For a better understanding definition of the terms "dynamic system identification" and "dynamic system" is given. The classification of methods for solving problems: analytical and compensatory; static and non-static; gradient and non-gradient; search and non-search. In more details described the methods associated with statistical data processing: ordinary least squares; generalized least squares; weighted least squares; maximum likelihood estimation; Bayesian methods; regularization methods. For some methods, there are images for clarity of work. It also explains why it is necessary to use these methods to identify the dynamic system and deliver the final result.