Excellnt, this works! Thanks!! ... View more

data dat2 missing_dat;   set dataset2;   if date=. then output missing_dat;   else output dat2; run; data want;   set dataset1      dat2; run; proc sort data=want;   by ID DATE PREFIX; run; data want_final;   set want      missing_dat; run; ... View more

SQL solution - data have; input id $3. log_code $3.; datalines; 11 A 12 A 12 B 13 A 14 A 14 B run; proc sql;   create table want as     select *    from have   where id not in (select distinct id      from have        where log_code in ('B')); quit; ... View more

Try this - data test1;   input building $3. type $8. content $8.; datalines; A22 Office chair A22 Office desk A22 Res.   bed A25 Res.   bed A25 Res.   lamp A26 Office chair A26 Res.   chair A26 Res.   bed A26 Res.   lamp run; proc sort data=test1;   by building type content; run; proc transpose data=test1 out=test1_wide;   by building type;   var content; run; data test2;   length content $30.;   set test1_wide;   retain building type content;     content = trim(col1)||' '||trim(col2)||' '||trim(col3);   keep building type content; run; ... View more

I feel better to do this in data step. ll giv you  more handle on achieving conditional statement. ... View more

Referencing the 2 steps I mentioned above shift the the date in one of the qtr var such that its matches w qtr of other one. That is one of the ways I look at it. ... View more

Try runngin this and view the the qtr1 var in both again. ll get it - data a3;   set locl.a3;   qtr1=qtr;   format qtr1 date9.; run; data a4;   set locl.a4;   qtr1=qtr;   format qtr1 date9.; run; QTR var in A3 and A4 actually ve unmatching valurs. you ll ve to change the var in either of the dataset first and then apply merge/join. ... View more

Try using this - class lev measure / style = {background = '#F1E5DB'} class lev measure / style = {background = '#F1E5DB'} class lev measure / style = {background = '#F1E5DB'} ... View more

There seem tp be something going on with the format for variable qtr in both the tables. Try running below step and have a look at qtr1 variable which is not formatted. ll ovbserve the difference. libname locl 'C:\test'; data a3;   set locl.a3;   qtr1=qtr;   *format qtr1 YYQ6.; run; data a4;   set locl.a4;   qtr1=qtr;   *format qtr1 YYQ6.; run; ... View more

Responding to your original code, looks like you would like to parse across 50 columns(diag) and thru 1000 rosw to set condition for Arrhythemia. Heres my quick solution. Giv it a shot - data diag_array;   set daig_cohort;   eligdiag = 0;   array diag(50) dx1-dx50;   do i=1 to 50; IF diag(i) EQ '427.31' THEN DO;   eligdiag = 1;      Arrhythemia='flu1'; END; ELSE IF diag EQ '427.32' THEN Arrhythemia='flu2' THEN DO;   eligdiag = 1;   Arrhythemia='flu2'; END; ELSE IF diag in('427','427.0','427.2','427.3','427.60', '427.61', '427.8', '427.81', '427.89', '427.9') THEN DO;   eligdiag = 1;   Arrhythemia='Other'; END;     ELSE DO;   eligdiag = 0;   Arrhythemia='None'; END;     IF eligdiag = 1 then leave;   end;   DROP eligdiag; run; the LEAVE statement is the key here. Gud Luk! ... View more