Thanks @dcruik and @PhilC. I did realize I have to include all the independent variables in for the VIF. Just couldn't get back yesterday to add a comment The explanations make sense @PhilC. I wanted something very precise to help me in my decision. This helped a lot. I will refer to the books suggested, Best, D R. ... View more

Thanks to you all. I really appreciate your information! ... View more

Yet another way: I read in Yes/no variables as 1 and 0 with custom informats. The Race0 = (Sum(raceA,raceB,raceP,raceI, raceW)>1); (I also tend to name such thins Race1 through 5 so Race1-Race5 makes better lists). ... View more

THank you!!! ... View more

You can generalize to any list of macro variables that have the property of the macro variable name matching the corresponding data set variable name. %let sera= ; %let kumulereignis=A,AA ; %let merkmal=BBB ; data _null_;   length where $30000 ;   array vars $500 sera kumulereignis merkmal ;   do i=1 to dim(vars);      vars(i)=symget(vname(vars(i)));      if vars(i) ne ' ' then         where =catx(' AND ',where,catx('=',vname(vars(i)),catq('1as',vars(i))))      ;   end;   if where ne ' ' then where='where ' || where ;   put where=;   call symputx('where',where); run; where=where kumulereignis='A,AA' AND merkmal='BBB' ... View more

I really appreciate your help. Also, thank you, dcruik, for your quick response, and different version to solve this problem. ... View more

Would this work? data want;    if _n_ = 1 then       do;          retain prx1-prx6;          prx1 = 's/100/Click/';          prx1 = 's/100/Click/';          prx2 = 's/117/Click2/';          prx3 = 's/121/Buy/';          prx4 = 's/123/Order/';          prx5 = 's/124/Click Out/';          prx6 = 's/000/Exit /';       end;    set have;    length Event_Path $ 1024;    Event_Path = Event_List;    Event_Path = prxchange(prx1, -1, Event_Path);    Event_Path = prxchange(prx2, -1, Event_Path);    Event_Path = prxchange(prx3, -1, Event_Path);    Event_Path = prxchange(prx4, -1, Event_Path);    Event_Path = prxchange(prx5, -1, Event_Path);    Event_Path = prxchange(prx6, -1, Event_Path);    drop prx1-prx6; run; ... View more

The first thing that came to mind is dcruik solution but here's another way.  Depending on what you have an need this might work: DATA TEST; A=1; B=2; output; RUN; data look_up; input lookup; cards; 1 2 3 4 5 ; run; proc sql; create table want as select a,5 as B from test where exists (select lookup           from look_up); ... View more

Appreciate your help everyone!! ... View more

Just to add another non-macro alternative (not that I think the macro language is a bad choice for this problem), I'm amazed by the magic of what can be done with the hash object.  Below hash-splitting is based on example from a Paul Dorfman paper, http://www.lexjansen.com/nesug/nesug09/hw/HW04.pdf 219  *data _null_ step that outputs multiple datatsets ; 220  data _null_  ; 221    if _n_=1 then do ; 222      declare hash h() ; 223      h.definekey ("_n_") ; 224      h.definedata ("store","product","date","buysell","units","price") ; 225      h.definedone () ; 226    end; 227    do _n_ = 1 by 1 until (last.store) ; 228      set have ; 229      by store ; 230      h.add() ; 231    end ; 232    h.output (dataset: cats("Store_",store) ) ; 233    h.clear() ; 234  run ; NOTE: The data set WORK.STORE_1 has 4 observations and 6 variables. NOTE: The data set WORK.STORE_2 has 5 observations and 6 variables. NOTE: There were 9 observations read from the data set WORK.HAVE. ... View more