The HPBIN Procedure

INPUT Statement

INPUT variables </ option> ;

The INPUT statement names one or more variables as input variables for binning. The specified variables must be interval variables. If classification variables are provided, PROC HPBIN stops with an error message. PROC HPBIN does not support duplicate variables. If the INPUT statement contains a duplicate variable, PROC HPBIN takes only the first variable and provides a warning message. You can specify the following option in each INPUT statement:

NUMBIN=integer

specifies the number of binning levels for all the binning variables in the current INPUT statement. The value of integer can be any integer between 2 and 1,000, inclusive.

The resulting number of binning levels might be less than the specified integer if the sample size is small or if the data are not normalized. In these cases, PROC HPBIN provides a warning message.

The number of binning levels that you specify in an INPUT statement overwrites the global number of binning levels. If you do not specify the NUMBIN= option in the INPUT statement, PROC HPBIN uses the global number of binning levels, which is 16 by default but can be specified in the NUMBIN= option in the PROC HPBIN statement.

When a BINS_META data set is specified, PROC HPBIN does not do binning and ignores the INPUT statement. Instead, PROC HPBIN takes the binning results from the BINS_META data set and calculates the weight of evidence and information value for the interval variables in the BINS_META data set.