Difference of a value from one variable and the lead of another value in the same variable??

[sandhya88]

Here is how my sample data looks like

ID$ Tst$ test_rt number

1    dp    20          1

1    rp     0            1

1    dp    35          2

1    rp     20          2

1    dp     0           3

1    rp      20         3

I would like to have an algorithm setup for calculating the difference between dp at number 1 and rp at value at next number?? Did any one encounter such situations, If so how to tackle that?? Please suggest, Thanks in advance

[Jagadishkatam]

Please try

data have;
input ID$ Tst$ test_rt number;
cards;
1 dp 20 1
1 rp 0 1
1 dp 35 2
1 rp 20 2
1 dp 0 3
1 rp 20 3
;

 
data want;
set have;
by number notsorted;
retain new_number;
if first.number then new_number=test_rt;
if last.number then dif=test_rt -new_number;
run;

[mkeintz]

By

---

"difference between dp at number 1 and rp at value at next number".

---

I think you want the rp test value minus the dp test value for each number.  If so, then:

data want;
  merge  have (where=(tst='dp') rename=test_rt=dp_test)
         have (where=(tst='rp') rename=test_rt=rp_test);
  by number;
  rp_minus_dp=rp_test-dp_test;
  drop tst ;
  if last.number;
run;

If that's not what you mean, please show what the results would look like for the data you provided.