GDA Introductory Overview - Comparison with Other Stepwise Discriminant Analysis Programs
STATISTICA General Discriminant Analysis (GDA) Models provides a unique, highly flexible application of the general linear model to the analysis of classification problems. Specifically, GDA's implementation permits you to build models for highly complex designs, including designs with effects for categorical predictor variables. Thus, the "general" in General Discriminant Analysis's refers both to the use of the general linear model, and to the fact that unlike most other stepwise discriminant analysis programs, GDA is not limited to the analysis of designs that contain only continuous predictor variables. (However, please refer to the Note of Caution for Models with Categorical Predictors, and Other Advanced Techniques to learn about the possible limitations of this approach.)
GDA is a "sister program" to the comprehensive STATISTICA General Linear Models and General Regression Models modules. Both modules provide similar methods for specifying analyses and producing results, so learning how to use one module makes it very easy to use the other. With the exception of options for using the overparameterized model (hence GDA ONLY uses the sigma-restricted model) and options for analyzing incomplete designs, all the innovative features of GLM are also available in GDA. The GDA - Unique features topic highlights only some of the unique features of the GDA module, that are usually not found in other (less complete) programs for performing discriminant analysis (refer also to Comparison with Other General Linear Model Programs).