SUPERLEARNER Procedure

INPUT Statement

INPUT variables </option>;

The INPUT statement specifies one or more predictor variables for the super learner model that share a common variable type. If your analysis has predictor variables of different types (for example, interval variables and nominal variables), you must specify multiple INPUT statements, one for each type. If you specify multiple INPUT statements, the union of variables that you specify in all the INPUT statements is used as predictor variables for each base learner model. However, if you specify any input-options in a BASELEARNER statement, then that base learner no longer uses variables that you specify in the INPUT statements as predictors.

You can specify the following option after a slash (/):

LEVEL=value

specifies the variable type. You can specify the following values:

INTERVAL: specifies that the variables are interval, which must be numeric.
NOMINAL: specifies that the variables are nominal, also known as classification variables, which can be numeric or character.

By default, LEVEL=INTERVAL for numeric variables and LEVEL=NOMINAL for categorical character variables.

Last updated: June 22, 2026