If you want to know the unique set of values BY variables assume for each cross section in the data file, use the OUTBY= option. For example, the following statements list some of the cross sections available for an IFS file, and are shown in Figure 11.3.

   filename ifsfile "imfifs1.dat" RECFM=F LRECL=88;
   proc datasource
      filetype=imfifsp infile=ifsfile
      outselect=on ebcdic
      interval=month
      outby=xsection;
   run;
   
   title1 'Some Cross Sections Available in IFSFILE';
   proc print data=xsection;
   run;

The OUTBY= data set reports the total number of series, NSERIES, defined in each cross section, NSELECT of which represent the selected variables. If you want to see the descriptive information on each of these NSELECT variables for each cross section, specify the OUTALL= option. For example, the following statements print descriptive information on all monthly series defined for all cross sections (COUNTRY=’111’, COUNTRY=’112’, COUNTRY=’146’, COUNTRY=’158’, and COUNTRY=’186’) which are shown in Figure 11.4.

   filename datafile "imfifs1.dat" RECFM=F LRECL=88;
   
   title3 'Time Series Defined in Cross Section';
   proc datasource filetype=imfifsp
      outselect=on ebcdic
      interval=month
      outall=ifsall;
   run;
   
   title4 'Cross Sections Available in OUTALL=IFSALL Data Set';
   proc print
      data=ifsall;
   run;

The OUTCONT= data set contains one observation for each time series variable with the descriptive information summarized over BY groups. When the data file contains no cross sections, the OUTCONT= and OUTALL= data sets are equivalent, except that the OUTALL= data set also reports time ranges of available data. The OUTBY= data set in this case contains a single observation reporting the number of series and time ranges for the whole data file.