BART Procedure
The PARTITION statement specifies how to logically partition observations in the input data set into disjoint subsets for model training and testing. For more information, see the section Using Validation and Test Data in Chapter 2, Shared Concepts. Either you can designate a variable in the input data table and a set of formatted values of that variable to determine the role of each observation, or you can specify proportions to use for randomly assigning observations to each role.
You must specify exactly one of the following partition-options:
-
FRACTION(<TEST=fraction><SEED=number>)
randomly assigns specified proportions of the observations in the input data table to the roles. You specify the proportions for testing by using the TEST= suboption. The specified fraction must be between 0 and 1. The remaining fraction is the proportion of the observations that are assigned to the training role. The SEED= option specifies an integer that is used to start the pseudorandom number generator for random partitioning of data for training, testing, and validation. If you omit the SEED= option, or if the SEED= option value is less than or equal to 0, the seed is generated by reading the time of day from the computer’s clock.
-
ROLE=variable(<TEST='value'><TRAIN='value'>)
ROLEVAR=variable(<TEST='value'><TRAIN='value'>)
names the variable in the input data table whose values are used to assign roles to each observation. You cannot also use this variable as an analysis variable in other statements or options. The TEST= and TRAIN= suboptions specify the formatted values of this variable that are used to assign observation roles. If you omit the TRAIN= suboption, then all observations whose role is not determined by the TEST= suboption are assigned to the training role. If you omit both the TEST= and TRAIN= suboptions, data partitioning is not performed.
Last updated: May 14, 2026