Content area
Full Text
Time Series Analysis: Forecasting and Control, 5th edition, George E. P. Box, Gwilynm M. Jenkins, Gregory Reinsel, and Greta M. Ljung. John Wiley & Sons, Ltd., Hoboken, NJ, 2016, 712 pp., $150.00, ISBN: 978-1-118-67502-1
A STATISTICIAN or data analyst may need a refresher reading on time series given a number collaborative projects from the engineering sciences. Such a person would excitedly await their copy of this new edition of the classic time series text. Certainly, a thorough and well developed theoretical exposition of the Box-Jenkins approach to time series modeling would be expected from these authors. The reader will find that the text does not disappoint in this respect. However, what really sets this edition apart from previous versions, is the contribution from the newly added author, Dr. Greta M. Ljung, to incorporate code from the open source statistical software program, R. This allows the reader put theoretical results and examples into action. Readers will surely profit from being able to recreate the results from the text and experimenting with variations of the codes that are provided.
The material for this edition follows that of its predecessor quite closely. There is a clear emphasis on ARIMA modeling using the Box-Jenkins approach as the focus of the first eight chapters. As indicated in the preface and in the review by Woodward (2009) for the fourth edition, a substantial amount of new material had been added to that edition which included nonlinear and long memory models (Chapter 10) and multivariate time series (Chapter 14). In the...