Multivariate Adaptive Regression Splines (MARSplines) Startup Panel - Quick Tab

The Quick tab of the Multivariate Adaptive Regression Splines (MARSplines) Startup Panel contains options to select variables and their types for the analysis. If you select the dependent variables as categorical, MARSplines will treat the problem as a classification task; otherwise it is regression. You can also exclude certain data cases of your choice using the Response codes and Predictor codes buttons (see below). Refer also to the Introductory Overview and Technical Notes.

Element Name Description
Variables Click the Variables button to display a standard variable selection dialog box. Select one or more continuous dependent variables, and a list of predictor variables. Because of the nature of MARSplines models, you can specify continuous and/or categorical predictor variables for the analysis. A single analysis will be performed for all variables, fitting common basis functions with different coefficients for each dependent variable. See also the Introductory Overview for additional details.
Response codes Click this button to display the Select Codes for Dependent Variables dialog box, which contains options to select the codes identifying the levels for the dependent variable. Codes must be integer values or text labels (and can be dates, times, etc.); at least two codes must be specified for the categorical dependent variable.
Predictor codes Click this button to display the Select Codes for Independent Variables dialog box, which contains options to select the codes identifying the levels for the predictor (independent) variables. Codes must be integer values or text labels (and can be dates, times, etc.); at least two codes must be specified for the categorical independent variables.