Works like a charm. Thanks. ... View more

Hi, The select returns about 4 million records. I am interested in finding records that exist in both datasets without dropping them with no visibility. Any idea how will the merge function. Thanks ... View more

Thanks guys, I have the solution. So instead of looking and my primary sorting variable, i used a variable that changes. 1033194759 1033194752 PAG 01JAN2001 E Anytime 001 1033194759 1033194752 TEL 01JAN2001 E Anytime 001 1033194759 1033194752 TEL 01JAN2001 E Daytime Only 002 data want;   set have;   by var1 var2 var3;   format CONSNUMBER $3.;   retain count;   if first.var3 then   count=1;   else count=count+1;   CONSNUMBER = put(count,z3.);   ; run; ... View more

Thanks guys. Just picked up something after implementing the code you all suggested. See below: 1033162957 1033162952 TEL 01JAN2001 E Anytime 001 1033162957 1033162952 TEL 01JAN2001 E Daytime Only 002 1033162957 1033162952 PAG 01JAN2001 E Anytime 003 The last record has a value PAG in the third field and I will like it to remain 001. In essecnce, the sequence will be 001,002,001 because the field is different in a way. If say we had 2 of each record, that is 2 records with TEL and 2 records with PAG then we should have 001,002,001,002. Sorry I was not aware of this case. Thanks a lot guys. ... View more