The HPDECIDE Procedure

PROC HPDECIDE Statement

PROC HPDECIDE <options>;

The PROC HPDECIDE statement invokes the procedure and identifies the input and output data sets. You also need the following statements:

DECISION statement
either a POSTERIORS or a PREDICTED statement
TARGET statement

You can specify the following options in the PROC HPDECIDE statement:

DATA=<libref.>SAS-data-set

specifies the input data set that contains the output from a modeling procedure. The default is the most recently created data set. If the data are already distributed, PROC HPDECIDE reads the data alongside the distributed database. For information about the various execution modes and about the alongside-the-database model, see the sections Processing Modes and Alongside-the-Database Execution in Chapter 2: Shared Concepts and Topics.

OUT=<libref.>SAS-data-set

specifies the output data set, which always contains any variables from the input data set that is specified in the ID statement, the chosen decision (with a prefix of D_), and the expected consequence of the chosen decision (with a prefix of either EL_ or EP_).

If the target value is in the input data set, then the output data set also contains the following variables: the consequence of the chosen decision (which is computed from the target value and has a prefix of either CL_ or CP_) and the consequence of the best possible decision when the target value is known (this variable has a prefix of either BL_ or BP_).

If the profit matrix is revenue, then the output data set also contains the following variables: the investment cost (which has a prefix of IC_) and the return on investment (which has a prefix of ROI_).

Additionally, if the PRIORVAR= and OLDPRIORVAR= variables are specified, then this data set contains the recalculated posterior probabilities.

OUTSTAT=<libref.>SAS-data-set

specifies an output data set to contain fit statistics. These statistics include the total and average profit or loss. By default, this data set is not created.

ROLE=TRAIN | VALID | VALIDATION | TEST | SCORE

specifies the role of the input data set. This option affects the variables that are created in the DATA= data set. The default value is TRAIN or SCORE. You can specify the following values:

TRAIN: specifies that the role of the input data set is training.
VALIDATION: specifies that the role of the input data set is validation.
TEST: specifies that the role of the input data set is testing.
SCORE: specifies that the role of the input data set is scoring.

Note: The selection of ROLE does not change the values in the OUTSTAT data set. If the role of the input data set is validation, letter "V" is added to the variable names of the total and average profit or loss. If the role of the input data set is testing, letter "T" is added to the variable names of the total and average profit or loss.

Last updated: January 04, 2019