This book is focused on modeling of linear dynamic systems. These models can be continuous-time or discrete-time. In both cases, the model can be written as a rational form in a generalized frequency variable. The aim of the book is to identify the parameters of these rational forms starting from noisy measurements of the input and output signals. The complete identification process is covered from starting from the experiment design, the parameter estimation-model selection, and the model validation. Since all real life systems behave to some extent nonlinearly, the influence of the nonlinearities on the linear idenfication framework is also studied in detail. The methods are illustrated on a very wide variety of practical examples coming from the electrical, mechanical, acoustical, and chemical fields. A great deal of attention will be spent on the practical applicability of the identification techniques. For clarity of exposition, the emphasis will be put on single input, single output systems. Most the presented results are also valid for multivariate systems and where necessary, the subtle differences will be explained.