BART Procedure

MODEL Statement

MODEL response=variables </ model-options>;

The MODEL statement specifies the response (dependent or target) variable and the predictor (independent or explanatory) variables. You can specify only one MODEL statement.

You can specify the following model-options:

DIST=keyword DISTRIBUTION=keyword

specifies the response distribution for the model. The keywords and their associated distributions are shown in Table 2.

Table 2: Built-In Distribution Functions

	Distribution
DISTRIBUTION=	Function
BINARY \| BERNOULLI	Binary
NORMAL \| GAUSSIAN \| GAUSS	Normal

If you do not specify a response distribution, the BART procedure determines the response distribution as follows. If you specify a GAUSSPRIOR statement, then a Gaussian response model is used. If the response variable is a character variable or included in the CLASS statement, then a binary response model is used. A Gaussian response model is used by default if none of these conditions are met.

OFFSET=variable

specifies a variable to use as an offset for the binary response model. For more information about the BART model for binary response data, see the section BART Model for Binary Data.

Last updated: June 22, 2026