If you are subtracting beginning price from end price here you go:   data have; infile cards dsd; input Cusip$ Months$ price; cards; 111111,1985 Aug,2.05 111111,,, 111111,1990 Jul,5.36 100000,1985 Aug,5.8 100000,, 100000,1990 Jul,8.6 ; data prep; set have; if not missing(months) then do; _month = month(input(put(cats('01',scan(months,2,' '),2015),$9.),date9.)); end; _year = scan(months,1,' '); run; proc sort data=prep out=prep2;by cusip _year _month;where not missing(months); data want; set prep2; retain _begin; by cusip _year _month; if first.cusip then _begin = price; if last.cusip then _end = price; return = _end-_begin; drop _:; run; ... View more

IML code: Code: Program data have; call streaminit(1234); do i=1 to 100;   a=ceil(rand('uniform')*100);   b=ceil(rand('uniform')*100);   output; end; drop i; run; %let low=0.05; %let high=0.95; proc iml; use have; read all var _num_ into x[c=vname]; close have; call qntl(q,x,{&low , &high}); do i=1 to ncol(x); x[loc(x[,i]<q[1,i]),i]=.; x[loc(x[,i]>q[2,i]),i]=.; end; create want from x[c=vname]; append from x; close want; quit; ... View more

I have the time right now so I'm taking a guess at what you are looking for.  An example of what you are looking for would be helpful. Try this: data have; infile cards; input Year Firm $ Scores; cards; 2005 A 5 2005 B 8 2006 A 7 2006 C 4 2007 A 8 2007 B 10 2007 D 8 2007 E 3 ; run; proc sql; create table prep as select distinct year, sum(scores) as sum_scores from have group by year order by year; data start(drop=l_sum_scores); set prep; by year; l_sum_scores = lag(sum_scores); diff_scores = abs(l_sum_scores - sum_scores); run; proc sql; create table want as select *,mean(diff_scores) as mean_scores from start; ... View more