Hi, You can create the macro variable which contains all the columns from dataset one and dataset two...by this way you can modify 5th line of my code... proc contents data = one noprint               out  = one_name(keep = name where = (name ^in ("Cusip" "Year"))); run; proc contents data = one noprint               out  = one_name(keep = name where = (name ^in ("Cusip" "Year"))); run; proc sql noprint;    select name into :one_list separated by ","    from one_name;    select name into two_list separated by ","    from two_list; quit; And after that just use the following line insted of 5th line... &one_list., &two_list., -Urvish ... View more

Hi Ksharp, Thanks for correction...Here i have updated my above code...Hope it works in case of any kind of Data Structure... Data one; input Cusip Year aa; cards; 100 97 10 101 98 2 102 97 . 103 98 2 ; Data two; input Cusip Year bb; cards; 101 98 . 102 97 . 103 98 3 104 99 5 ; proc sql;    create table both as    select coalesce(a.cusip,b.cusip) as cusip,          coalesce(a.year,b.year) as year,          aa,bb,          case                  when a.cusip = b.cusip then "Match"                  when aa = . and bb = . then "Match"              when bb = . then "Two"              when aa = . then "One"              else "Match"          end as source    from one as a        full join        two as b    on a.cusip = b.cusip; quit; -Urvish ... View more

Use the %DO Iterative Loops within the macro to refrence many variables at one time...If you are not cleared about no of variables in your dataset then count the no of variables first of all and then save it in one macro variable...I have assumed that there are 20 variables in your dataset... I have not tested it but should use...Hope it works... %macro obs_indentify(no_of_vars =);    proc sql;       create table both as      select coalesce(a.cusip,b.cusip) as cusip,             coalesce(a.year,b.year) as year,             aa,bb,             case                                  %do i = 1 %to &no_of_vars.;                    when bb&i. = . then "Two"                    when aa&i. = . then "One"                                %end; else "Match"           end as source        from one as a                   full join         two as b        on a.cusip = b.cusip;     quit; %mend; %obs_identify(no_of_vars = 20); -Urvish ... View more

Can you please post the sample SAS Code to let the other people know about why did you get the unexpected output...??? I don't think it is BLACK HOLE ERROR in SAS if it is then SAS Stops to respond...It will be better if you post sample code... -Urvish ... View more

You can follow the SAS/SQL dictionary or user's manual... http://support.sas.com/documentation/cdl/en/sqlproc/62086/PDF/default/sqlproc.pdf Enjoy... ... View more

Hi, Try with the following code...Although it is hard coded at one place to recreate the flags,it meets your requirement... data have;     input student $ test $ flag1 - flag3;       cards4; Carl A 1 0 0 Carl B 0 1 0 Tito A 1 0 0 Tito B 0 1 0 Tito C 0 0 1 ;;;; proc sql noprint;    create table have1 as    select student,count(student) as no_of_student    from have    group by 1;    select max(no_of_student) into :max      from have1; quit; data want(drop = i);    set have1;    array flag(&max.);    do i = 1 to &max.;       flag(i) = 1;         if no_of_student < &max. then do;       flag3 = 0;         end;         /*Based on your requirement, add more flags*/    end; run; -Urvish ... View more

Try following code... data want;    input data_elaboration      8.          filler1             $1.          ch                  $1.          filler2             $1.          sw                  $1.          filler3             $1.          data_initialization 8.          filler4             $1.          data_fin            8          filler5             $1.          hour_ini            8.          filler6             $1.          hour_fin            8.          filler7             $1. @@;       cards4; 12345678 C T 12358789 12547898 20145789 4578912 ;;;; Hope it helps... -Urvish ... View more