The PROC HPREDUCE statement invokes the procedure. Table 1 summarizes the important options in the PROC HPREDUCE statement by function. The options are then described fully in alphabetical order.
You can specify the following options in the PROC HPREDUCE statement.
CORR
selects variables based on the correlation matrix. Assuming that X and Y are two variables, the correlation between X and Y is computed by:
This is the default option for computing the Pearson correlation statistics in PROC HPREDUCE.
COV
selects variables based on the covariance matrix. Assuming that X and Y are two variables, the covariance between X and Y is computed by:
DATA=SAS-data-set
names the input SAS data set to be used by PROC HPREDUCE. The default is the most recently created data set. If the procedure executes in distributed mode, the input data are distributed to memory on the appliance nodes and analyzed in parallel, unless the data are already distributed in the appliance database. When data are already distributed, the procedure reads the data alongside the distributed database. See the sections Processing Modes and Alongside-the-Database Execution in Chapter 2, Shared Concepts and Topics.
FMTLIBXML=file-ref
specifies the file reference for the XML stream that contains the user-defined format definitions. In a distributed computing environment, user-defined formats are handled differently than they are in other SAS products. For information about how to generate an XML stream for your formats, see the section Working with Formats in Chapter 2, Shared Concepts and Topics.
NAMELEN=number
specifies the length to which long effect names are shortened (20 number128). The default and minimum value is 64.
NOCLPRINT<=number>
suppresses the display of the "Class Level Information" table if you do not specify number. If you specify number, the values of the classification variables are displayed for only those variables whose number of levels is less than number. Specifying a number helps to reduce the size of the "Class Level Information" table if some classification variables have a large number of levels.
NOPRINT
suppresses the generation of ODS output.
NOSUMPRINT
suppresses the generation of the "Selection Summary" table.
OUTCP=SAS-data-set</LIST<(EPS =number)>>
creates both a data set that contains a symmetric matrix that depicts the relationships among variables and also a set of statistics about the input data set and variables. Depending on the Pearson correlation statistics option specified in the PROC HPREDUCE statement, the symmetric matrix can be a correlation (CORR) matrix, a covariance (COV) matrix, or a sums of squares and crossproducts (SSCP) matrix.
When the LIST option is specified, the symmetric matrix is output in the list-of-list (LIL) format. In this format, the matrix is represented as a set of tuples , where x is an entry in the matrix and i and j denote its row and column indices, respectively. LIL format can be used when the output contains too many columns to fit in a data set. For example, in most database systems the maximum number of columns in a table is usually limited to several thousand. If an output matrix contains more columns than the limit, you must use the LIST option to avoid errors that would arise from writing too many columns to the table. When LIL format is used, all 0 entries in the matrix are ignored in the output.
When EPS= number is specified in the LIST option, matrix entries that have an absolute value smaller than number are ignored in the output. This feature helps omit unreliable estimations and generate a compact representation for the matrix. When the EPS= option is not specified, only the 0 entries in the matrix are ignored in the output.
SSCP
selects variables based on the sums of squares and crossproducts matrix. Assuming that X and Y are two variables and that and are their corresponding variable vectors, the SSCP between X and Y is computed by:
TECHNIQUE=keyword
TECH=keyword
specifies the variable selection technique. You can specify the following keywords:
VARIANCEANALYSIS | VAR
performs variance analysis for variable selection.
DISCRIMINANTANALYSIS | DSC
performs discriminant analysis for variable selection.
The default value is TECHNIQUE=VAR.
You can use variance analysis for both supervised and unsupervised variable selection. You can use discriminant analysis only for supervised variable selection with one classification variable as the response. For more information, see the section Variable Selection for Classification.
TIMEPRINT
prints the time (in seconds) used by each variable selection iteration in the "Selection Summary" table. If this option does not appear in the PROC HPREDUCE statement, the time information is not printed.