SAS for Forecasting Time Series

John C. Brocklebank (Author), David A. Dickey (Author)

9780471395669, Wiley

Paperback / softback, published 7 February 2006

424 pages, Tables: 76 B&W, 0 Color; Graphs: 59 B&W, 0 Color
27.9 x 21.1 x 2.3 cm, 0.989 kg

"The new material and the update of the excellent 1E, now 17 years in the past, certainly make the 2E a necessary purchase for any user of SAS time series modeling methods."
Technometrics
Vol. 46, No. 1, February 2004

Easy-to-read and comprehensive, this book shows how the SAS System performs multivariate time series analysis and features the advanced SAS procedures STATSPACE, ARIMA, and SPECTRA. The interrelationship of SAS/ETS procedures is demonstrated with an accompanying discussion of how the choice of a procedure depends on the data to be analysed and the reults desired. Other topics covered include detecting sinusoidal components in time series models and performing bivariate corr-spectral analysis and comparing the results with the standard transfer function methodology. The authors? unique approach to integrating students in a variety of disciplines and industries. Emphasis is on correct interpretation of output to draw meaningful conclusions. The volume, co-pubished by SAS and JWS, features both theory and practicality, and accompanies a soon-to-be extensive library of SAS hands-on manuals in a multitude of statistical areas. The book can be used with a number of hardware-specific computing machines including CMS, Mac, MVS, Opem VMS Alpha, Opmen VMS VAX, OS/390, OS/2, UNIX, and Windows.

Chapter 1- Overview of Time Series.

Chapter 2- Simple Models: Autoregression.

Chapter 3- The General ARIMA Model.

Chapter 4- The ARIMA Model: Introductory Applications.

Chapter 5- The ARIMA Model: Special Applications.

Chapter 6- State Space Modeling.

Chapter 7- Spectral Analysis.

Subject Areas: Mathematics [PB]