General Discriminant Analysis (GDA) Models Startup Panel and Quick Tab

Ribbon bar. Select the Statistics tab. In the Advanced/Multivariate group, click Mult/Exploratory and from the menu, select General Discriminant to display the General Discriminant Analysis (GDA) Models Startup Panel.

Classic menus. On the Statistics - Multivariate Exploratory Techniques submenu, select General Discriminant Analysis Models to display the General Discriminant Analysis (GDA) Models Startup Panel.

The Startup Panel contains one tab: Quick. Use the options on this tab to select the Type of analysis and Specification method.

See also the General Discriminant Analysis (GDA) Models Index, Overviews, and Examples.

As described in the Introductory Overview, the General Discriminant Analysis (GDA) Models module of STATISTICA provides an extension to the traditional approach to discriminant function analysis (e.g., as described in the context of the Discriminant Analysis module) by (dummy-)coding the categorical dependent variable values (i.e., class or group memberships) into multiple dependent variables. STATISTICA then analyzes the data as a multivariate regression problem, as described in the Introductory Overview of the General Regression Models (GRM) module. While traditionally discriminant function analysis is typically applied to data files with single degree of freedom continuous predictor variables, with the GDA module, you can specify complex models involving continuous and categorical predictor variables and effects, and perform stepwise and best-subset selection of those predictors. However, refer to the Note of caution for models with categorical predictors, and other advanced techniques, to learn about the possible limitations of this approach.

Type of analysis

The Types of Analyses list presents two choices for the type of discriminant analysis model (see the Introductory Overview). Select the type of design that you want to perform.

Traditional discriminant analysis

Select Traditional discriminant analysis to specify analyses involving only single degree of freedom continuous predictor variables (see also the Introductory Overview). Note that this is the "traditional" discriminant analysis as also implemented in the Discriminant Analysis module; however, in addition GDA provides options to perform best subset selection of predictors (as well as stepwise selection, which is also supported in the Discriminant Analysis module). On the subsequent Quick specs dialog, you can then specify one or more continuous predictor variables, and one categorical dependent variable.

General discriminant analysis

Select General discriminant analysis to specify ANOVA- and ANCOVA-like designs for continuous and categorical predictor variables and effects. The types of designs that can be specified are analogous to those supported by the General Linear Models module (see the Introductory Overview of the STATISTICA General Linear Models module). On subsequent dialogs, you can specify forward and backward stepwise selection of predictor effects, as well as best subset selection of effects. Like the analogous options available in the General Regression Models module, when performing stepwise and best-subset selection of predictor effects, multiple-degree of freedom effects will not be broken into single degree of freedom predictors.

Most, if not all, textbook treatments of discriminant function analysis are limited to simple and stepwise analyses with single degree of freedom continuous predictors. No "experience" (in the literature) exists regarding issues of robustness and effectiveness of these techniques, when they are generalized in the manner provided in this very powerful module. Refer to the Note of caution for models with categorical predictors, and other advanced techniques to learn about the possible limitations of this approach.

Specification method

The Specification method list presents the three alternative user interfaces available in GDA (as well as STATISTICA General Linear Models, General Regression Models, Generalized Linear/Nonlinear Models, and General Partial Least Squares Models). You can choose among the three different user interfaces in the Specification method list only when General discriminant analysis is selected as the Type of analysis. When Traditional discriminant analysis is selected, you must use Quick specs dialog.

Quick specs dialog

Select Quick specs dialog to display the respective Quick Specs dialog box corresponding to the current selection in the Type of analysis list. The Quick Specs - Quick tab will prompt you to select a categorical dependent variable and categorical and/or continuous predictor variables (depending on the selection in the Type of analysis box), and construct a default model. Use the options on the Advanced tab to modify various computational specifications, or the Syntax editor button in the Quick specs dialog box to further customize the model via command syntax (see GDA Syntax).

Analysis Wizard

Select Analysis Wizard to display a sequence of dialog boxes that will guide you through the steps for specifying an analysis that may involve a complex design for the predictor variables. At the conclusion of the sequence of dialog boxes, you can either compute the results, or click the Syntax editor button in the GDA Models Wizard Extended Options dialog box to further customize the model via command syntax, open an existing file with command syntax, or save syntax in a file for future repetitive use. Note that selecting Analysis Wizard automatically moves the Type of analysis selection to General discriminant analysis. You cannot use the Analysis Wizard with any other type of analysis.

Analysis syntax editor

Select Analysis syntax editor to specify a model via the GDA Models Syntax Editor. The GDA Models Syntax Editor provides various options for specifying designs and for modifying various parameters of the computational procedures. You can also open an existing text file with command syntax, or save syntax in a file for future repetitive use. Refer to the description of GDA Syntax for addition details. Note that selecting Analysis syntax editor automatically moves the Type of analysis selection to General discriminant analysis. You cannot use the Analysis syntax editor with any other type of analysis.

Note. Refer to the Introductory Overview for details concerning analyses in GDA that involve categorical and/or continuous predictor effects, in an ANOVA-like design.

OK. Click the OK button to display the appropriate analysis specification dialog box, depending on the Specification method selected on the Quick tab.

Cancel. Click the Cancel button to close the Startup Panel without performing an analysis.

Options. See Options Menu for descriptions of the commands on this menu.

Open Data. Click the Open Data button to display the Select Data Source dialog box, which contains options to choose the spreadsheet on which to perform the analysis. The Select Data Source dialog box contains a list of the spreadsheets that are currently active.

Select Cases. Click the Select Cases button to display the Analysis/Graph Case Selection Conditions dialog box, which contains options to create conditions for which cases will be included (or excluded) in the current analysis. More information is available in the case selection conditions overview, syntax summary, and dialog description.

W. Click the W (Weight) button to display the Analysis/Graph Case Weights dialog box, which contains options to adjust the contribution of individual cases to the outcome of the current analysis by "weighting" those cases in proportion to the values of a selected variable.

Weighted moments. Select the Weighted moments check box to specify that each observation contributes the weighting variable's value for that observation. The weight values need not be integers. This module can use fractional case weights in most computations. Some other modules use case weights as integer case multipliers or frequency values. This option will only be available after you have defined a weight variable via the W option (see above).

DF = W-1 / N-1

When the Weighted moments check box is selected, moment statistics (e.g., mean, variance) can be based on the sum of the weight values for the weighting variable (W-1), or on the number of (unweighted) observations (N-1). The sums of squares and cross products will always be based on the weighted values of the respective observations. However, in computations requiring the degrees of freedom (e.g., standard deviation, ANOVA tables), the value for the degrees of freedom can either be computed based on the sum of the weight values, or based on the number of observations. Moment statistics are based on the sum of the weight values for the weighting variable if the W-1 option button is selected, and are based on the number of (unweighted) observations if the N-1 options button is selected. For more information on options for using integer case weights, see also Define weight.

See also the General Discriminant Analysis (GDA) Models Index, Overviews, and Examples.