John Wiley & Sons Inc SAS Forecasting 2e

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. Review "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 From the Inside Flap In this second edition of the indispensable SAS® for Forecasting Time Series, Brocklebank and Dickey show you how SAS performs univariate and multivariate time series analysis. Taking a tutorial approach, the authors focus on the procedures that most effectively bring results: the advanced procedures ARIMA, SPECTRA, STRATESPACE, and VARMAX. They demonstrate the interrelationship of SAS/ETS ® procedures with a discussion of how the choice of a procedure depends on the data to be analyzed and the results desired. With this book, you will learn to model and forecast simple autoregressive and vector ARMA processes using the STATE-SPACE and VARMAX procedures. Other topics covered include detecting sinusoidal components in time series models, performing bivariate cross-spectral analysis, and comparing these frequency-based results with the time domain transfer function methodology. New and updated examples in the second edition include Retail sales with seasonality ARCH models for stock prices with changing volatility Vector autoregression and cointegration models Intervention analysis for product recall data Expanded discussion of unit root tests and nonstationarity Expanded discussion of frequency domain analysis and cycles in data Data mining and forecasting with examples using SAS IntelliVisor Using the HPF procedure to automatically generate forecasts for several time series in one step From the Back Cover In this second edition of the indispensable SAS® for Forecasting Time Series, Brocklebank and Dickey show you how SAS performs univariate and multivariate time series analysis. Taking a tutorial approach, the authors focus on the procedures that most effectively bring results: the advanced procedures ARIMA, SPECTRA, STRATESPACE, and VARMAX. They demonstrate the interrelationship of SAS/ETS ® procedures with a discussion of how the choice of a procedure depends on the data to be analyzed and the results desired. With this book, you will learn to model and forecast simple autoregressive and vector ARMA processes using the STATE-SPACE and VARMAX procedures. Other topics covered include detecting sinusoidal components in time series models, performing bivariate cross-spectral analysis, and comparing these frequency-based results with the time domain transfer function methodology. New and updated examples in the second edition include Retail sales with seasonality ARCH models for stock prices with changing volatility Vector autoregression and cointegration models Intervention analysis for product re