The HPCANDISC Procedure

OUT= Data Set

Many SAS procedures add the variables from the input data set when an observationwise output data set is created. The assumption of high-performance analytics procedures is that the input data sets can be large and can contain many variables. For performance reasons, the OUT= data set contains the following:

  • new variables that are explicitly created for the OUT= data set

  • variables that are listed in the ID statement

  • distribution keys or hash keys that are transferred from the input data set

Having these variables and keys in the OUT= data set enables you to add output data set information that is necessary for subsequent SQL joins without copying the entire input data set to the output data set. For more information about output data sets that are produced when PROC HPCANDISC is run in distributed mode, see the section Output Data Sets.

The new variables that are created for the OUT= data set contain the canonical variable scores. You determine the number of new variables by using the NCAN= option. The names of the new variables are formed as they are for the PREFIX= option. The new variables have means equal to 0 and pooled within-class variances equal to 1.