The HPPRINCOMP Procedure

Output Data Sets

When an observationwise output data set is created, many SAS procedures add the variables from the input data set to the output data set. High-performance statistical procedures assume that the input data sets can be large and can contain many variables. For performance reasons, the output data set contains only the following:

  • variables that are explicitly created by the statement

  • variables that are listed in the ID statement

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

Including these variables and keys 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 you run PROC HPPRINCOMP in distributed mode, see the section Output Data Sets in SAS/STAT 14.1 User's Guide: High-Performance Procedures.