Skip to content

Make Validation Column

The Make Validation Column utility enables JMP standard users to easily construct a validation column that defines a test set. JMP Pro users have a way to construct validation columns (Analyze > Predictive Modeling > Make Validation Column), but previously JMP standard users did not. This utility enables JMP standard users to construct a validation column that selects a specified proportion of test rows randomly.

To construct a validation column in your data table using this utility:

  1. Open the Predictum menu > Self-Validating Ensemble Modeling > Utilities > Make Validation Column option.

  2. In the Test Proportion text box, specify the value of your desired proportion of test set rows.

  3. In the Random Seed field, enter a random seed value, if you want to be able to repeat this construction in the future.

    Validation Column Dialog

  4. Click OK.

    A new Validation column is added as the last column in the data table. This column has a nominal modeling type and the value in each row is indicated as either “Training” or “Test”.

    Added Validation Column

If you specified a random seed value, it is saved as a Notes property for the Validation column.