Latent Variable Models and Factor Analysis

representation of the Volterra model. One of the main issues addressed in Chapter 6 is the criterion by which local models are selected. These models will be designated as input, output, or general selected. The author, in this chapter, also presents similarities and dissimilarities between input-selected multimodels and NMAX, output-selected and NARX, and general selected and NARMAX. Chapter 7 focuses on the relationships that exist between these different model classes. Finally, Chapter 8 concludes by emphasizing the four steps—selection, physical phenomenon, goodness of Ž t, and assessment of the validity of the model. Overall, the author’s presentation is insightful and consistent. It is well written and nicely organized. The book’s strength is its verbal explanation of the mathematical deŽ nitions and results; intuitive discussions and applications from various physical Ž elds are frequently supplied. From the very beginning of the book, the author goes to a great length to explain the material to ensure that the reader understands the differences among the models discussed. The book’s appeal is that it illustrates a wide range of results for many kinds of models that appear in stochastic processes and time series literature. It provides general insights on the nonlinear model classes that have been discussed in the modeling and control literature. Finally, this book is an important addition in the area of nonlinear time series, and it has much to offer that is hard to Ž nd elsewhere.