ABSTRACT
This chapter presents to the fundamental methods of empirical model building. It introduces the least squares concept and deriving its central formula for simple static systems. The chapter defines the polynomial then to general rational structures. It discusses the basic off-line algorithm can be converted into on-line form, with various degrees of recursivity. The chapter discusses the unbiased methods goes far beyond the scope. It summarizes the standard foundations of systems identification methodology, somewhat re-emphasized and re-interpreted based on the author's teaching and research experience. The chapter is devoted to the existence criteria of the estimation algorithm. It investigates the effect of noise on the accuracy of the estimates. The chapter deals with the problems and techniques of selecting the model structure. It derives the fundamental least squares parameter estimation algorithm, for a simple static linear system. The chapter extends to nonlinear static systems which are linear in the parameters.
