The SCA System provides many new capabilities and enhancements that are reorganized into packages from Educational to Advanced Editions: Time series power transformation analysis and diagnostics Improved forecasting using power transformations Time-varying parameter models Generalized threshold AR and ARIMA modeling Segmented time series modeling and forecasting GARCH modeling and application environment New seasonal ARIMA identification method Unit root testing Causality tests using vector ARIMA models Improved estimation with root checking of ARMA factors Date building, handling, indexing, and aggregation Educational Edition (Academic Users) The SCA Educational Edition includes essential time series analysis and forecasting capabilities for teaching and learning. It is this fundamental module on which other SCA forecasting and time series products are built. The Educational Edition focuses on time-tested modeling capabilities, providing all the necessary tools to identify, estimate, diagnostically check, and forecast using various time series models. Box-Jenkins nonseasonal/seasonal ARIMA modeling New identification method for seasonal ARIMA models Powerful transfer function modeling and forecasting Effective LTF model identification for transfer functions Lagged regression with autocorrelated errors Intervention (impact) analysis Exponential smoothing using various methods Time series simulation Constrained parameter estimation A wide array of capabilities for general statistical analysis Large workspace (60,000 words) allocation Practitioner Edition The Practitioner Edition builds on the Education Edition by adding expert-system automatic time series modeling and forecasting features. It is easy to use and is an asset to novices and experts alike. In addition, it includes power packed capabilities to automatically detect and adjust for outliers during estimation which is a great tool for time series data mining. The Practitioner Edition offers an effective solution to handle repetitive modeling and forecasting tasks on a large number of time series. It is a natural choice in driving large scale forecasting applications that require automation. Automatic identification of seasonal and nonseasonal ARIMA models Automatic transfer function modeling and intervention analysis Automatic detection and adjustment for outliers using a joint estimation algorithm by C. Chen and L.-M. Liu Automatically handles level shifts, temporary changes, additive, and innovational outliers Improved estimates of intervention effects through joint estimation of model parameters and outlier effects Better forecasting results by special handling of outliers occurring at the end of a time series Time series data mining and exploration Improve forecasting using power transformation Model identification and estimation with missing data Trading day and moving holiday adjustment Date functions to facilitate daily, weekly, and monthly modeling and forecasting Unrestricted workspace allocation Professional Edition (A) The Professional Edition (A) builds on the Practitioner Edition by adding multivariate time series analysis and forecasting using vector ARIMA and simultaneous transfer function (STF) models. These advanced modeling approaches are well-suited to business, economic, industrial and social science time series data. Analyze and forecast multivariate time series using vector ARIMA models Causality tests using vector ARIMA models Analyze and forecast multivariate time series using simultaneous transfer function models that accommodate for intervention, trading day and moving holiday effects Ideal for the analysis of spatial time series data Multivariate time series simulation using vector ARIMA and STF models Study contemporaneous effects using structural form models employing STF models, or use reduced form vector ARMA models to leverage lagged dependencies Extend upon conventional econometric models by addressing serially correlated errors

Professional Edition (B) The Professional Edition (B) builds on the Practitioner Edition by adding power transformation, segmented time series methods, nonlinear time series testing, identification, modeling, forecasting using TAR, threshold ARIMA, and threshold transfer function models. Also includes new analysis capabilities for time-varying parameter models and GARCH models.

New criterion for power transformation to improve forecasting accuracy Segmented time series modeling and forecasting using weighted estimation methods Effective handling of clustered outliers, and desensitizing parameter estimates from temporary structural changes Threshold autogressive (TAR) and general threshold ARIMA modeling and forecasting Piecewise and threshold transfer function modeling and forecasting Nonlinearity tests on time series Analysis of time-varying parameter models GARCH modeling (See SCAB34S GARCH below) Model-based seasonal decomposition Advanced Edition The Advanced Edition provides SCA’s full breadth of capabilities to model and forecast time series data combining the features of Professional Editions (A) and (B). SCAB34S GARCH ARCH, GARCH, GARCH-M, Integrated GARCH, Exponential GARCH, Threshold GARCH, GJR GARCH, GARCH with regressors, Bivariate GARCH, and more Normal, student-t, Standard Cauchy, and General Error Distributions Advanced diagnostic statistics and graphics Generalized nonlinear estimation for custom models Comprehensive applet environment providing flexible statistical and mathematical programming. Direct two-way interface with other SCA products SCAB34S SPLINES Multivariate Adaptive Regression Splines modeling Adaptive Threshold Autoregression (ASTAR) modeling for time series General Additive Models (GAM) Alternating Conditional Expectation (ACE) modeling Simplified model results easily adapted into external scoring systems Contribution charts and leverage charts Predictive modeling and database scoring capability Advanced diagnostic statistics and graphics Direct two-way interface with other SCA products B34S ProSeries Econometric System
The B34S® ProSeries Econometric System is a comprehensive software system for econometric modeling and analysis. Among its many features are:

Includes SCAB34S GARCH and SPLINES products
PI-spline, ACE, and GAM non-linear modeling
State space models and non-linear testing of models
Kalman filtering
Regression estimation using OLS, GLS, L1 and MINIMAX models
BLUS and RA analysis, error component analysis, and recursive residual analysis
Principal component analysis and QR model estimation
LOGIT, PROBIT, TOBIT
Spectral analysis
Comprehensive applet development environment and matrix programming language Basic System Features
Compatible with Windows Vista, XP, 2000, NT
Programmable command language
Interactive menu and batch execution environments
Import data from spreadsheets and databases
Automate tasks using SCA macro procedures
Group a large number of data and analysis tasks into a project file and execute the project as a batch operation
Graphical menus and dialogs for all SCA commands
Tabulate time series model information from multiple SCA sessions into consolidated summary tables
Graphical user interface providing easy-to-use application environments for SCAB34S GARCH, SCAB34S SPLINES, and other SCA components