Lynn Nicole Lethbridge <lynnl@is.dal.ca> wrote: : Does anyone know of an efficient way to do subtraction across observations : in SAS. For example suppose I have the following data: : ID var1 var2 : 1 4 7 : 1 3 8 : 2 6 10 : 2 5 7 : I'd like to subtract var1 from var1 in the previous observation and var2 : from var2 in the previous observation BUT ONLY WHERE THERE ARE 2 IDS. So : for ID=4 or ID=7 there will be no subtraction.

use a sort followed with a data step which uses set and by statements. then use the first./last. syntax to identify where new idss begin. then, rather than using the lag function and then subtracting, (as others have suggested), use the DIFn(var) function to compute the n'th laggin difference.

steve