ok, 10x. ... View more

Hello, A data step solution: data have; input a b c d e; datalines; 1 3 5 7 9 2 4 6 8 10 ; data want; set have; array vars{*} a--e; do i=1 to dim(vars); character=vname(vars{i}); number=vars{i}; output; end; keep character number; run; ... View more

hello, probably it is because of LAG(CLAIM_DATE), executed conditionally. if sometimes the condition is not true the results are not the expected ones. ... View more

Hello, This macro is a starting point. there are some notes in the log about the invalid argument of the INPUT function, but there will cause no harm. /*prepare some data*/ data a_20131212 a_20131213 a_20131214 a_20140112; set sashelp.class; run; %macro delt(lib, inpdate, daysbehinddel); %let refdate=%sysfunc(intnx(day,&inpdate, %eval((-1)*&daysbehinddel))); proc sql noprint; select memname into :deltables separated by " " from dictionary.tables where not(missing(input(scan(memname,2,"_"),yymmdd8.))) and  input(scan(memname,2,"_"),yymmdd8.) lt %eval(&refdate) and upcase(libname)=upcase("&lib"); quit; %put &deltables; proc datasets library=&lib nolist; delete &deltables; quit; %mend delt; /*example of calling the macro*/ %delt(work,"01JAN2014"d,5) ... View more

hello, a data step solution: data have; input Date ddmmyy10. Time :time5. library $ Var1 Var2; format date date9. time time5.; datalines; 09/01/2015 00:00 xxxxx 2 6 09/01/2015 00:15 xxxxx 1 9 09/01/2015 00:30 xxxxx 5 8 09/01/2015 00:45 xxxxx 0 10 09/01/2015 01:00 xxxxx 7 6 09/01/2015 01:15 xxxxx 9 6 09/01/2015 01:30 xxxxx 10 0 09/01/2015 01:45 xxxxx 11 9 09/01/2015 02:00 xxxxx 18 10 10/01/2015 03:00 xxxxx 7 6 10/01/2015 03:15 xxxxx 9 6 10/01/2015 03:30 xxxxx 10 0 10/01/2015 03:45 xxxxx 11 9 10/01/2015 04:45 xxxxx 12 19 ; run; proc sort data=have; by date; run; data want (keep=dateout hourinterval maxvar1 maxvar2) ; set have end=last; by date; retain dateout hourflag maxvar1 maxvar2; format dateout date9.; if first.date then do;   if _N_ ne 1 then do;*write to output when first date changes;    hourinterval=cats(hourflag||"-"||left(hour(time)));    output;   end;   dateout=date;   maxvar1=var1;   maxvar2=var2;   hourflag=hour(time);   put hourflag=; end; else if hourflag=hour(time) then do;   if maxvar1 lt var1 then maxvar1=var1;   if maxvar2 lt var2 then maxvar2=var2; end; else if hourflag ne hour(time) then do;   put _N_=;   hourinterval=cats(hourflag||"-"||left(hour(time)));   output;   hourflag=hour(time);   maxvar1=var1;   maxvar2=var2; end; /*write last record*/ if last then do;   hourinterval=cats(hourflag||"-"||left(hour(time)));   output; end; run; ... View more

hello, just to show a different approach: data have; input statistics $ Treatment1 Treatment2 Treatment3 Treatment4 ; datalines; Mean 3.4 0 2 0 Min-Max 0 0 0 0 N 2 0 0 0 SD 0 0 0 0 ; proc means data=have; output out=r (drop=_type_ _Freq_) sum= / autoname; run; %let nvar=%sysfunc(attrn(%sysfunc(open(r,i)),nvars)); %put &nvar; data want; retain Treatment:; set have; if _n_=1 then set r; array treat{*} Treatment: ; array newtreat{&nvar} $ newtreat1-newtreat4; do i=1 to &nvar; if treat{&nvar+i}=0 then newtreat{i}='Na'; else newtreat{i}=left(treat{i}); end; drop i Treatment:; run; ... View more

Hello, With proc transpose: data test; input guest_id tran_id amount; datalines; 101 11230 499 102 11254 566 101 11245 699 103 11345 999 102 12145 799 103 11564 699 101 11897 245 102 12456 600 ; run; proc sort data=test out=testsorted; by guest_id; run; proc transpose data=testsorted out=tr1; by guest_id ; var tran_id amount ; run; proc transpose data=tr1 out=want (drop=_name_); by guest_id; id _name_; var col2; run; ... View more

hello, data have; input ID Var1 Var2 Var3  ; datalines; 1 20 50 30 2 0.90 0.25 0.10 ; run; data want; set have; array vars{*} var1-var3; max_val=max(var1, var2,var3); min_val=min(var1, var2,var3); do i=1 to dim(vars) until(vars{i}=max_val); WANT_Highest=vname(vars{i}); end; do i=1 to dim(vars) until(vars{i}=min_val); WANT_Lowest=vname(vars{i}); end; drop max_val min_val i; run; ... View more

Hello, proc sql noprint; select name into :vars separated by ' ' from sashelp.vcolumn where memname='CLASS' and libname='SASHELP' AND name like '%e'; quit; %put &vars; data want; set sashelp.class (drop=&vars) ; run; ... View more