Enter your mobile number or email address below and we'll send you a link to download the free Kindle Reading App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your email address or mobile phone number.
It is too bad that this book was not published by one of the major publishers of statistics books. It is a very important but largely overlooked treatise on model selection procedures for regression and time series. The authors are experts in this area who have made important contributions to the theory. There is a wealth of techniques for model selection and there has been much confusion about their properties and usefulness. This book covers most of the methods, is very much up-to-date and clears up much of the confusion right in the first chapter! Techniques for univariate regression, autoregressive time series models, multivariate regression, vector autoregression, cross-validation, bootstrap, robust regression, nonparametric regression and wavelets are all covered. Many practical examples are given to illustrate the methods and there are also a number of useful simulation studies that appear in the book. The final chapter (Chapter 9) covers extensive simulations comparing many of the popular model selection criteria for both time series and regresion modeling.
My only disappointment is the omission of the recent developments in Bayesian model selection. At least the authors mention this omission upfront in Chapter 1 and provide good references to the literature.
Was this review helpful to you?