Example of Stepwise Variable Selection
In this example, you use stepwise variable selection in the Discriminant platform to determine which variables are included in the final model.
1. Select Help > Sample Data Folder and open Iris.jmp.
2. Select Analyze > Multivariate Methods > Discriminant.
3. Select Sepal length, Sepal width, Petal length, and Petal width and click Y, Covariates.
4. Select Species and click X, Categories.
5. Select Stepwise Variable Selection.
6. Click OK.
7. Click Step Forward three times.
Figure 5.15 Stepped Model for Iris.jmp
Three covariates are entered into the model. The Smallest P to Enter appears in the top panel. It is 0.0103288, indicating that the remaining covariate, Sepal length, might also be valuable in a discriminant analysis model for species.
8. Click Apply This Model.
Figure 5.16 Score Summaries Report Showing Selected Covariates
The window is updated to show reports for a fit based on the entered covariates and your selected discriminant method. Note that the covariates that you selected for your model are listed at the top of the Score Summaries report.