topic Update only the desired number of observations in SAS Programming

Update only the desired number of observations

Santt0sh — Tue, 28 Jun 2022 04:27:43 GMT

Hi All!!

I am trying to create a variable based on the values of another variable in the same dataset.
Dataset original has 137 records and after calculations and applying various conditions the values of the Variable FLAG are 1 and 0 the sum of Variable FLAG is 29. Now I am trying to create a new variable TOT_FLAG
Where the TOT_FLAG is only has the value 1 for 29 observations and 0 for the remaining 108 observations.
I have tried using the do until for some reason my code isn't updating only 29 observations.

Please suggest.

I'm not able to post the code now but will try again in sometime

Regards,
S

Re: Update only the desired number of observations

Kurt_Bremser — Tue, 28 Jun 2022 05:15:52 GMT

Use _N_:

data want;
set have;
tot_flag = (_n_ le 29);
run;

Re: Update only the desired number of observations

Santt0sh — Tue, 28 Jun 2022 12:49:40 GMT

Hi Kurt,

Thank you for your quick response!!

I'm sorry I forgot to mention that I need to update all the 29 observations out of 137 observations where another Variable "TRANSACTIONS" is 1
and the remaining 108 observations needs to be 0.
I was trying the below code, the issue seems to be with _N_ here
As it will only go till _N_ = 29 but I am trying to update the 29 observations where Transactions = 1 out of 137 observations, and the remaining 108 Flag needs to be assigned 0.

DATA Transactions;
Set Transactions;
If Transactions EQ 1 and _N_ LE 29 then do;
Total_Flag = 1;
END;
ELSE Do;
TOTAL_FLAG = 0;
END;
Run;

Kindly Suggest!

Re: Update only the desired number of observations

Kurt_Bremser — Tue, 28 Jun 2022 14:19:20 GMT

So we can not use the data step iteration counter, but need to make our own:

data want;
set have;
retain __n 1;
if __n lt 29 and transactions eq 1
then do;
  total_flag = 1;
  __n + 1;
end;
else total_flag = 0;
drop __n;
run;

Re: Update only the desired number of observations

Santt0sh — Tue, 28 Jun 2022 14:44:14 GMT

Thank you Kurt!!!
I worked, to update all the 29 observations I had to change LT to LE.

Regards,
S