@ilikesas wrote: Hi Reeza,   why did you do in the code  do j=1 to ncomb+1; instead of  "do j=1 to ncomb;"    I ran the code on a small sample data: data have; input A B C D E F; datalines; 1 2 3 4 5 6 10 20 30 40 50 60 ; run;   and noticed that the 21st combination for each original observation is the exact copy of the 20th combination (and in this case ncomb = 20)? I didn't check that logic to be honest, the OPs initial code used it and I left it in, but clearly that doesn't make sense. It should probably be ncomb.  ... View more