Loading......

    This book is aimed at the reader who wishes to gain a working knowledge of timeseries and forecasting methods as applied in economics, engineering and the naturaland social sciences. Unlike our earlier book, Time Series: Theory and Methods, re-ferred to in the text as TSTM, this one requires only a knowledge of basic calculus,matrix algebra and elementary statistics at the level (for example) of Mendenhall,Wackerly and Scheaffer (1990). It is intended for upper-level undergraduate studentsand beginning graduate students.
    The emphasis is on methods and the analysis of data sets. The student versionof the time series package ITSM2000, enabling the reader to reproduce most of thecalculations in the text (and to analyze further data sets of the reader's own choosing),is included on the CD-ROM which accompanies the book. The data sets used in thebook are also included. The package requires an IBM-compatible PC operating underWindows 95, NT version 4.0, or a later version of either of these operating systems.The program ITS M can be run directly from the CD-ROM or installed on a hard diskas described at the beginning of Appendix D, where a detailed introduction to thepackage is provided
    Very little prior familiarity with computing is required in order to use the computerpackage. Detailed instructions for its use are found in the on-line help files whichare accessed, when the program ITSM is running, by selecting the menu optionHelp>Contents and selecting the topic of interest. Under the heading Data youwill find information concerning the data sets stored on the CD-ROM. The book canalso be used in conjunction with other computer packages for handling time series.Chapter 14 of the book by Venables and Ripley (1994) describes how to performmany of the calculations using S-plus.
    There are numerous problems at the end of each chapter, many of which involveuse of the programs to study the data sets provided.
    To make the underlying theory accessible to a wider audience, we have statedsome of the key mathematical results without proof, but have attempted to ensurethat the logical structure of the development is otherwise complete. (References toproofs are provided for the interested reader.)
    Since the upgrade to ITSM2000 occurred after the first edition of this bookappeared, we have taken the opportunity, in this edition, to coordinate the text withthe new software, to make a number of corrections pointed out by readers of the firstedition and to expand on several of the topics treated only briefly in the first edition.
    Appendix D, the software tutorial, has been rewritten in order to be compatiblewith the new version of the software.
    Some of the other extensive changes occur in (i) Section 6.6, which highlightsthe role of the innovations algorithm in generalized least squares and maximumlikelihood estimation of regression models with time series errors, (ii) Section 6.4,where the treatment of forecast functions for ARIMA processes has been expandedand (iii) Section 10.3, which now includes GARCH modeling and simulation, topicsof considerable importance in the analysis of financial time series. The new materialhas been incorporated into the accompanying software, to which we have also addedthe option Autof~t. This streamlines the modeling of time series data by fittingmaximum likelihood ARMA(p, q) models for a specified range of (p, q) values andautomatically selecting the model with smallest AICC value.
    There is sufficient material here for a full-year introduction to univariate and mul-tivariate time series and forecasting. Chapters 1 through 6 have been used for severalyears in introductory one-semester courses in univariate time series at Colorado StateUniversity and Royal Melbourne Institute of Technology. The chapter on spectralanalysis can be excluded without loss of continuity by readers who are so inclined.
    We are greatly indebted to the readers of the first edition and especially to MatthewCalder, coauthor of the new computer package, and Anthony B rockwell for theirmany valuable comments and suggestions. We also wish to thank Colorado StateUniversity, the National Science Foundation, Springer-Verlag and our families fortheir continuing support during the preparation of this second edition.Fort Collins, Colorado
   
 
   
                                                                                                                      Peter J. Brockwell  August 2001
                                                                                                                      Richard A. Davis