Thank you ... View more

Read my post again, thoroughly. In particular note how output is directed to datasets. With your data step, all records are written to both datasets, and so you get missing values in half of them, when variables are not read by an input statement. ... View more

data req; input id$ salary; datalines; 001 2000 002 3000 003 4000 001 3000 001 5000 004 2000 003 1000 002 5000 004 1000 ; run; proc sql; select * from (select * from req except (select * from req group by id having salary=max(salary))) group by id having salary=max(salary); quit; ... View more

@PeterClemmensen: @rvsidhu035 already has accepted a solution but, of course, I agree with you. In this case I, personally, would use the macro after running proc transpose to make the dataset long. i.e.: data new30; input id $ sysbp diabp; cards; 001 130 80 002 110 90 003 140 70 004 120 80 005 130 70 ; run; proc transpose data=new30 out=need; var sysbp diabp; by id; run; %transpose(data=need, out=want, by=_name_, id=id, prefix=_, var=col1, use_varname=no, sort=yes) Requires a lot less code, has a lot less chance of making an error, and should take a lot less time than all of those steps.   Art, CEO, AnalystFinder.com   ... View more

This is the wrong approach for this problem as there is a function, INDEXC(), that does what you need already.   But the reason you are getting multiple outputs is a logic problem with how you structured your DO loop.  You should not have an OUTPUT statement in the loop.  Instead you should use the loop to set the value.  Typically you set the initial value and then when you find a match change the value.  Note that DO statement allows you to have both an iterative part and a other parts.  So you can stop searching once you have found changed the destination value.   So you could construct your DESTINATION variable like this: length destination $15; destination='DOMESTIC'; do I=1 to length(dat) while (destination='DOMESTIC'); if substr(dat,I,1) in ('X','Y','Z') then destination='INTERNATIONAL'; end;   ... View more

K ... View more

data given; infile datalines dlm=' ' missover dsd; input subject visit $11. visitnum 16-18; datalines; 100 Screening -1 100 day 1 1 100 day 2 2 100 UNSCHEDULED 100 UNSCHEDULED 100 day 3 3 100 day 4 4 100 UNSCHEDULED 100 day 5 5 100 UNSCHEDULED 100 UNSCHEDULED 100 UNSCHEDULED ; run; data output1(rename=(nvar=visitnum)); length subject 3.; length visit $20.; set given; if visitnum ge 1 or nmiss(visitnum) eq 1 then do; if nmiss(visitnum) eq 1 then do; nvar+0.01;visit=catx(':',visit,put(nvar,4.2));end; else nvar=visitnum;visit=visit;end; else nvar=visitnum; drop visitnum; run; ... View more

I fixed the title -- though the question looks like a homework assignment... ... View more

data dsn; input city$ road$; cards; delhi a delhi b delhi c mumbai d mumbai a chennai e chennai f chennai a chennai g bangalore i bangalore j bangalore k bangalore a ; run; proc freq data=dsn; table road/out=dsn1;run; proc sort data=dsn1;by road;run; proc sort data=dsn;by road;run; data merg; merge dsn dsn1;by road;run; data want(keep=road city); set merg; if count=4;run; ... View more