GPREG Procedure

PROC GPREG Statement

PROC GPREG <options>;

The PROC GPREG statement invokes the procedure. Table 1 summarizes the options available in this statement.

Table 1: PROC GPREG Statement Options

Option	Description
Input and Output Data Set Option
DATA=	Specifies the input data table
Sparse Gaussian Process Options
FIXINDUCINGPOINTS	Specifies that the inducing points are to be fixed during optimization
NINDUCINGPOINTS=	Specifies the number of inducing points
Regression Output Options
OUTINDUCINGPOINTS=	Outputs the inducing points in the regression
OUTVARIATIONALCOV=	Outputs the covariance of the Gaussian process
Performance Options
NTHREADS=	Specifies the seed to start the random number generator
SEED=	Specifies seed for the computation

You can specify the following options:

DATA=libref.data-table

names the input data table for PROC GPREG to use. The default is the most recently created data table. libref.data-table is a two-level name, where

libref: refers to a collection of information that is defined in the LIBNAME statement and includes the library, which includes a path to the data. For more information about libref, see the section Using SAS Viya Workbench.
data-table: specifies the name of the input data table.

FIXINDUCINGPOINTS

specifies that the inducing points in the sparse Gaussian process are to be fixed or adjusted during the optimization. If you omit this option, PROC GPREG adjusts the locations of the inducing points during the iterations of the model inference.

NINDUCINGPOINTS=number

specifies the number of inducing points to use in the sparse Gaussian process. The number must be a positive integer. When you specify a larger number, the PROC GPREG provides a more precise prediction results for the regression, but at a slower processing speed. When you specify a smaller number, the result are less accurate but the processing is faster.

By default, NINDUCINGPOINTS=100.

NTHREADS=number-of-threads

specifies the number of threads to use in the computation. The default value is the number of CPUs available in the machine.

OUTINDUCINGPOINTS=libref.data-table

creates the data table that contains the inducing points to use in the sparse Gaussian process. If you also specify the FIXINDUCINGPOINTS option, this table contains the cluster centroids of the k-means used as the initial inducing points. Otherwise, the table contains the locations of the inducing points of the last iteration of the model inference. libref.data-table is a two-level name, where libref refers to the library, and data-table specifies the name of the output data table. For more information about this two-level name, see the DATA= option and the section Using SAS Viya Workbench.

OUTVARIATIONALCOV=libref.data-table

creates the data table that contains the covariance to use in the sparse Gaussian process. The table contains the covariance of the Gaussian process used in the last iteration of the model inference. libref.data-table is a two-level name, where libref refers to the library, and data-table specifies the name of the output data table. For more information about this two-level name, see the DATA= option and the section Using SAS Viya Workbench.

SEED=number

specifies an integer to be used to start the pseudorandom number generator. If you do not specify a seed or if you specify a value less than or equal to 0, the seed is generated by reading the time of day from the computer’s clock.

Last updated: September 23, 2026