Hi, Could you describe the problem in more detail? Maybe with a small example? Do you want to allocate marketing spends to different campaigns? Usually PROC OPTMODEL is the right tool to solve such problems. ... View more

Was not aware of you have done it yesterday ago, was still waiting on your new post. anyway, here is my suggested solution: data have;      infile cards dlm='09'x truncover;      input patient:$10.   (Visit_date     Max_date) (:ddmmyy10.)     Trt_at_visit$;      format visit_date Max_date ddmmyy10.;      cards; Patient1   01/12/00   30/05/01   Drug_x Patient1   31/12/00   29/06/01   Drug_x Patient1   24/03/01   20/09/01   Drug_e Patient1   22/07/01   18/01/02   Drug_x Patient1   02/12/00   31/05/01   Drug_x Patient1   01/01/01   30/06/01   Drug_x Patient1   25/03/01   21/09/01   Drug_x Patient1   23/07/01   19/01/02   Drug_x Patient1   20/11/01   19/05/02   Drug_a Patient1   20/12/01   18/06/02   Drug_x Patient1   13/03/02   09/09/02   Drug_x Patient1   11/07/02   07/01/03   Drug_x Patient1   08/11/02   07/05/03   Drug_x Patient1   08/12/02   06/06/03   Drug_x Patient1   01/03/03   28/08/03   Drug_x Patient2   29/06/03   26/12/03   Drug_x Patient2   27/10/03   24/04/04   Drug_e Patient2   26/11/03   24/05/04   Drug_x Patient2   17/02/04   15/08/04   Drug_x Patient2   16/06/04   13/12/04   Drug_x Patient2   14/10/04   12/04/05   Drug_e Patient2   13/11/04   12/05/05   Drug_x Patient2   04/02/05   03/08/05   Drug_x Patient2   04/06/05   01/12/05   Drug_x Patient2   02/10/05   31/03/06   Drug_x Patient2   01/11/05   30/04/06   Drug_x Patient2   23/01/06   22/07/06   Drug_x Patient2   23/05/06   19/11/06   Drug_x ; proc sql;      create table want as            select *, exists(select * from have where   patient=a.patient                 and a.visit_date <= visit_date <= a.visit_date+180                 and Trt_at_visit ne a.Trt_at_visit                 ) as Concurrent_flag      from have a      ; quit; Good luck, Haikuo ... View more