CATTRANSFORM Procedure

PROC CATTRANSFORM Statement

  • PROC CATTRANSFORM <options>;

The PROC CATTRANSFORM statement invokes the procedure.

You can specify the following options:

BINMISSING

places missing values into a separate bin whose bin ID value is 0.

DATA=libref.data-table

names the input data table for PROC CATTRANSFORM 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.

DISTINCTCOUNTLIMIT=integer

specifies the maximum level limit.

By default, DISTINCTCOUNTLIMIT=5000.

EVALUATIONSTATS(options)

requests evaluation statistics about the transformed variables. You can specify the following options:

WOEADJUST=number

specifies the weight-of-evidence adjustment value. The number can range from 0 to 1, inclusive.

By default, WOEADJUST=0.5.

WOEDEFINITION=EVENT | NONEVENT

specifies the weight-of-evidence definition. If WOEDEFINITION=EVENT, the weight of evidence is defined as event/nonevent. If WOEDEFINITION=NONEVENT, the weight of evidence is defined as nonevent/event.

By default, WOEDEFINITION=NONEVENT.

MISSINGEVALNONEVENT

treats missing targets as nonevents.

NBINS=integer

specifies the number of bins for all input variables.

OUTLEVELBINMAP=libref.data-table

saves the bin mapping information to a data table. The bin mapping table contains three columns: Variable, Level, and BinID. 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.

PREFIX="string"

specifies the prefix for the transformed variables.

By default, PREFIX="CAT".

PREPROCESSRARE

groups levels whose frequency percentages are 5% or less with the next least frequent level until all percentages are greater than 5% before implementing the specified method.

Last updated: May 14, 2026