Perfect, thank you! ... View more

Can you post an excerpt of the dataset you want to iterate over? I can't see the need to move date into macro-variables. Maybe using a hash object, dow-loops or even a format, but certainly no macro variables. ... View more

Is it correct that both obs with id=3 have month=1? This seems to work, too: data want; if 0 then set have; length r_flag l_flag 8; do _n_ = 1 by 1 until (last.id); set have; by id; if first.id or (sum(l_flag, r_flag) = 0) then do; r_flag = Right > Left; l_flag = Left > Right; end; end; do _n_ = 1 by 1 until (last.id); set have; by id; output; end; run;   ... View more

Best approach with modern SAS tools, requires no sorting: data match discrepancy ; set check; if _n_ = 1 then do; declare hash v (dataset:"valid"); v.definekey("id"); v.definedone(); end; if v.check() = 0 then output match; else output discrepancy; run; Untested; for tested code, provide usable example data in data steps with datalines. ... View more

data have; set sashelp.class; if _n_ in (1 4 7) then call missing(age, sex); run; %MACRO impute_freq (dat, var_name); proc freq data=&dat(where=(&var_name is not missing)) order=freq noprint; table &var_name/out=mode; run; data &dat; set &dat; if _n_=1 then set mode(keep=&var_name rename=(&var_name=_&var_name)); if missing(&var_name) then &var_name=_&var_name; drop _&var_name; run; %MEND impute_freq; %impute_freq(have,age) %impute_freq(have,sex) ... View more