You will need to create a new variable with the aggregated value of your three "comp" variables. See this SAS usage note with a great example. That should get you started and welcome to StackOverflow!
UPDATE: I think I better understand the problem: you need to denormalize your data (turn the columns into observations). So try this:
data test(keep=company value date);
set dev2.Actionfinal2;
company = 1; value = comp_1; output;
company = 2; value = comp_2; output;
company = 3; value = comp_3; output;
run;
proc sort data=test;
by date company;
run;
data test2;
set test;
by date;
if first.date then new_y=0;
new_y + value;
run;
symbol1 i=join v= cv= ci=red;
symbol2 i=join v= cv= ci=blue;
symbol3 i=join v= cv= ci=green;
proc gplot data=test2;
plot new_y*date=company
/ areas=3 vaxis=0 to 100000 by 20000;
run;
quit;