The chapter "Identification of Parametric Models from Experimental Data" provides a comprehensive overview of the methods and criteria used to identify parametric models from experimental data. It begins with an introduction to the aims of modeling, the system under study, and the model itself, emphasizing the importance of optimization and parameter uncertainty. The chapter then delves into various structural aspects of models, including phenomenological and behavioural models, linear and nonlinear models, continuous- and discrete-time models, deterministic and stochastic models, and the choice of model complexity. The criteria for model identification are discussed in detail, covering least squares, least modulus, maximum likelihood, Bayesian criteria, and robustness. The chapter also explores optimization techniques, such as LP structures, quadratic cost functions, least-squares estimators, and various optimization methods including gradient descent, Newton's method, and quasi-Newton methods. It addresses constrained optimization, non-differentiable cost functions, and global optimization techniques. The section on uncertainty addresses cost contours in parameter space, Monte-Carlo methods, and methods based on the density of the estimator. It also covers bounded-error set estimation and the representation of cost contours. The chapter concludes with a discussion on experiments, including criteria, local design, robust design, and the influence of model structure. Finally, it addresses the falsification of models through simple inspection and statistical analysis of residuals.The chapter "Identification of Parametric Models from Experimental Data" provides a comprehensive overview of the methods and criteria used to identify parametric models from experimental data. It begins with an introduction to the aims of modeling, the system under study, and the model itself, emphasizing the importance of optimization and parameter uncertainty. The chapter then delves into various structural aspects of models, including phenomenological and behavioural models, linear and nonlinear models, continuous- and discrete-time models, deterministic and stochastic models, and the choice of model complexity. The criteria for model identification are discussed in detail, covering least squares, least modulus, maximum likelihood, Bayesian criteria, and robustness. The chapter also explores optimization techniques, such as LP structures, quadratic cost functions, least-squares estimators, and various optimization methods including gradient descent, Newton's method, and quasi-Newton methods. It addresses constrained optimization, non-differentiable cost functions, and global optimization techniques. The section on uncertainty addresses cost contours in parameter space, Monte-Carlo methods, and methods based on the density of the estimator. It also covers bounded-error set estimation and the representation of cost contours. The chapter concludes with a discussion on experiments, including criteria, local design, robust design, and the influence of model structure. Finally, it addresses the falsification of models through simple inspection and statistical analysis of residuals.