Hi DN, your code is so nice! I added the RED part to your code. proc transpose data=counts prefix=CODE out=wanted_dataset;    by obs;    id comcode;    var _freq_;    run; ... View more

example: data have; input obs comcode; cards; 1 2 1 6 1 9 2 5 2 8 2 10 3 4 3 8 3 9 3 20 ; data want(drop=i); set have; by obs;   retain c1-c20 0; array _com(*) c1-c20; do i=1 to dim(_com); if comcode=i then _com(i)=1;      end; if last.obs then do;   output;   do i=1 to dim(_com);     _com(i)=0;    end;   end; proc print;run;         o               m               c        O  o   o                             c  c  c  c  c  c  c  c  c  c  c        b  b   d  c  c  c  c  c  c  c  c  c  1  1  1  1  1  1  1  1  1  1  2        s  s   e  1  2  3  4  5  6  7  8  9  0  1  2  3  4  5  6  7  8  9  0        1  1   9  0  1  0  0  0  1  0  0  1  0  0  0  0  0  0  0  0  0  0  0        2  2  10  0  0  0  0  1  0  0  1  0  1  0  0  0  0  0  0  0  0  0  0        3  3  20  0  0  0  1  0  0  0  1  1  0  0  0  0  0  0  0  0  0  0  1 ... View more

Hi PG, sorry,I am not wise enough to understand your question:smileysilly::smileyconfused::smileyblush:. You say " I would particularly like to get rid of the special missing value.", then you let value = .X;   /* Special missing value */. Don't you get a lot of .X after you run your code? data myData; update myData outliersUpdate; by date station; run; ... View more

Hi Peter, below is what I got by running your code. It is very nice. Thank you! data time0; input id $ var1-var5; cards; aa 5 4 3 2 1 bb 6 7 8 9 10 ; data time1; input id $ var1-var5; cards; aa 5 9 3 2 1 bb 8 7 8 9 11 ; proc sort data=time0;by id;run; proc sort data=time1;by id;run; proc compare base= time0 compare= time1 outall   out= compares noprint; id id ; run ; proc print data=compares label ; title time0 time1 ; id id _obs_ _type_ ; var var: ; run ;           Observation      Type of     id       Number      Observation      var1     var2    var3    var4    var5     aa         1           BASE              5.0000       4      3       2       1     aa         1           COMPARE       5.0000       9      3       2       1     aa         1           DIF                 0.0000       5      0       0       0     aa         1           PERCENT        0.0000     125    0       0       0     bb         2           BASE              6.0000       7      8       9      10     bb         2           COMPARE       8.0000       7      8       9      11     bb         2           DIF                 2.0000       0      0       0       1     bb         2           PERCENT      33.3333       0      0       0      10 ... View more

data test; input month $ sale; datalines; jan 2000 feb 1400 mar 2300 apr 3400 may 2700 jun 3200 jul 800 aug 2900 sep 1700 oct 2100 nov 2600 dec 3200 ; data want; set test; num_month=month(input(cats('01',month,'2013'),date9.)); run; proc print;run; ... View more

data class1 class2;   set sashelp.class; run; %macro macroa(vara,varb); data &vara._&varb;   set &vara &varb;   run; %mend; %macroa(class1,class2) proc print data=&syslast;run; ... View more

sample code: data time0; input id $ var1-var5; cards; aa 5 4 3 2 1 bb 6 7 8 9 10 ; data time1; input id $ var1-var5; cards; aa 5 9 3 2 1 bb 8 7 8 9 11 ; proc sort data=time0;by id;run; proc sort data=time1;by id;run; %let lib=work; %let dsn=time0; /* put your variables in a macro variable */ proc sql noprint;   select catx('=',name,cats(name,'_t1')) into : renames separated by '  '     from dictionary.columns   where libname="%upcase(&lib)" and memname="%upcase(&dsn)" and type='num' and lowcase(name) not in ('id')/* put   the variables you don't want to be calculated here */; select cats('diff_',name,'=',name,cats('-',name,'_t1')) into : diff separated by '; '     from dictionary.columns   where libname="%upcase(&lib)" and memname="%upcase(&dsn)" and type='num' and lowcase(name) not in ('id'); quit; /* rename your vareiables in dataset time1 */ proc datasets lib=work nolist;   modify time1;   rename &renames;   quit;   data want;    merge time0(in=a) time1(in=b);    by id;    if a and b;    &diff; run; proc print;run;                                                  d    d   d   d    d                                                           i    i   i   i    i                                     v   v   v   v    v    f    f   f   f    f                                     a   a   a   a    a    f    f   f   f    f                                     r   r   r   r    r    _    _   _   _    _                v   v   v   v    v   1   2   3   4    5    v    v   v   v    v       O        a   a   a   a    a   _   _   _   _    _    a    a   a   a    a       b   i    r   r   r   r    r   t   t   t   t    t    r    r   r   r    r       s   d    1   2   3   4    5   1   1   1   1    1    1    2   3   4    5       1   aa   5   4   3   2    1   5   9   3   2    1    0   -5   0   0    0       2   bb   6   7   8   9   10   8   7   8   9   11   -2    0   0   0   -1 ... View more

is this what you want? data have; input date mmddyy10.; format date mmddyy10.; cards; 01/20/2013 01/28/2013 02/20/2013 ; data want; set have; diff=dif(date); proc print;run; ... View more

Is the code helpful? proc sort data=dealers; by  id event_dt  descending size; run; data lib.dealers; set dealers; by  id event_dt; if first.event_dt then count=0; count+1; if count<3 then output; run; Message was edited by: Linlin ... View more