The trio of goals for this introductory time series book is to provide (1) mathematical completeness, (2) computational illustration and implementation, and (3) conciseness and accessibility to upper-level undergraduate and M.S. students. The emphasis on goal one has swamped the other goals to a degree. For example, the mathematical completeness squeezes undergraduates away from using the book. The book has flexibility, but one must pay close attention to the preface as how to use the book. Easy and difficult exercises are marked as well as theoretical and computational exercises. There are more than 600 exercises, half of which involve R coding and/or data analysis. Supplements include a website with 12 key datasets and all R code for the book’s examples, as well as the solutions to exercises. Many will like the use of R in that the coding will ingrain time series principles.
The book has 12 chapters: (1) Introduction, (2) The Probabilistic Structure of Time Series, (3) Trends, Seasonality, and Filtering, (4) The Geometry of Random Variables, (5) ARMA Models With White Noise Residuals, (6) Time Series in the Frequency Domain, (7) The Spectral Representation, (8) Information and Entropy, (9) Statistical Estimation, (10) Fitting Time Series Models, (11) Nonlinear Time Series Analysis (ARCH and GARCH covered here), and (12) The Bootstrap. There are also five appendices on topics of probability, mathematical statistics, asymptotics, Fourier series, and Stieltjes integration, which could be good refreshers for many students; but students needing some reminders on such topics would probably have problems with this book. For sure, no time series class would cover all the chapter topics so this book provides that flexibility of choice and most would skip Chapters 7 and 8.
I found that the best chapters of the book were the Introduction (for visualization, core topics, and coding) and the Bootstrap. The authors did a good job of pulling together much of what is happening currently with bootstrapping in time series. In my sphere of influence with companies and students, this book would be an excellent resource. It would have been nice to have seen at least one example worked out in each chapter. The other pragmatic suggestion would have been a chapter on outliers and missing data. There are theoretical and practical solutions to such difficulties. Depending upon who your client or your audience is, different options may be needed. Lastly, most of the data that I deal with has seasonality issues and not much on the topic.
I always have high expectations on time series books, but it is a tough topic with so many topics and issues. I commend the authors for their good attempt, which will be a stepping stone for other such books in a practical or applied way.
The University of Tennessee (Retired)