This generic tab offers a number of options to partition your data into learn, test and validate samples. It also includes cross-validation as an alternative.
There are a number of different test method available as:
-No independent testing- Results in an exploratory model with no independent testing. Resubstitution is used instead to approximate a test sample.
-Fraction of cases selected at random- Specifies a fraction of cases selected at random for testing and optionally, validation
-Test sample contained in a separate file- The test sample is contained in a separate dataset.
-Cross-validation -V fold cross-validation- Request number of folds for example 2 or 5 or 20
-Cross-validation-Variables determined CV bins- Specify variable which assigns records to cross-validation bins.
-Variable separates learn, test (holdout)- Specify variable which separates learn and test samples.