This book provides a comprehensive overview of parametric model identification from experimental data. It covers the aims of modeling, system and model definitions, and the criteria for model selection. The text discusses various model structures, including phenomenological, linear, nonlinear, continuous- and discrete-time models, as well as deterministic and stochastic models. It explores the concept of model complexity and structural properties, such as identifiability and distinguishability, with detailed methods for assessing these properties. The book then delves into different criteria for model selection, including least squares, least modulus, maximum likelihood, and Bayesian criteria. It addresses parameter constraints and robustness to noise uncertainty, along with methods for tuning hyperparameters. The optimization section covers various techniques, including linear programming, quadratic programming, gradient methods, and constrained optimization approaches. It also discusses optimization of measured responses and the use of response-surface methodology. The uncertainty section examines cost contours in parameter space, Monte-Carlo methods, density-based estimation, and bounded-error set estimation. The experiments section covers design criteria, local and robust design methods, Bayesian estimation, and the influence of model structure. Finally, the book discusses model falsification through simple inspection and statistical analysis of residuals, concluding with references and an index. The book is a detailed resource for understanding the principles and methods of parametric model identification from experimental data, providing a thorough treatment of the subject for researchers and practitioners in engineering and related fields.This book provides a comprehensive overview of parametric model identification from experimental data. It covers the aims of modeling, system and model definitions, and the criteria for model selection. The text discusses various model structures, including phenomenological, linear, nonlinear, continuous- and discrete-time models, as well as deterministic and stochastic models. It explores the concept of model complexity and structural properties, such as identifiability and distinguishability, with detailed methods for assessing these properties. The book then delves into different criteria for model selection, including least squares, least modulus, maximum likelihood, and Bayesian criteria. It addresses parameter constraints and robustness to noise uncertainty, along with methods for tuning hyperparameters. The optimization section covers various techniques, including linear programming, quadratic programming, gradient methods, and constrained optimization approaches. It also discusses optimization of measured responses and the use of response-surface methodology. The uncertainty section examines cost contours in parameter space, Monte-Carlo methods, density-based estimation, and bounded-error set estimation. The experiments section covers design criteria, local and robust design methods, Bayesian estimation, and the influence of model structure. Finally, the book discusses model falsification through simple inspection and statistical analysis of residuals, concluding with references and an index. The book is a detailed resource for understanding the principles and methods of parametric model identification from experimental data, providing a thorough treatment of the subject for researchers and practitioners in engineering and related fields.