Assigning unique ID to a subset of observations

[adityal2810]

I have a panel data with 1002 observations. N=334, T=3. My goal is to create 2 columns representing N and the year T. For example, the first 334 observations will have ID from 1 to 334 and a fixed value of T=1. The next set will have observations from 335 to 668 and T=2. I tried this piece of code but it did not work. 

DATA WAGE;
SET WAGE(first_obs = 2)(obs = 335);
retain Person_ID = 1;
retain Year_ID = 1;
Person_ID + 1;
RUN;

[PGStats · Accepted solution]

Could be as basic as:

data want;
do T = 1 to 3;
    do ID = 1 to 334;
        set WAGE;
        output;
        end;
    end;
run;

[ballardw]

Perhaps:

data want;
   set wage;
   id=_n_;
   t = floor((_n_-1)/334)+1;
run;