keep specific number of time series observations for each cross section

[AbuChowdhury]

Hi Experts,

In my dataset, there are a lot of time series observations for each firm. I would like to keep first 25 observations for each firm. How can I do that?

[PeterClemmensen]

Try this

data test;
   do _N_ = 1 by 1 until (last.stock);
      set sashelp.stocks;
      by stock;
      if _N_ <= 25 then output;
   end;
run;

[AbuChowdhury]

Hi,

The code keeps only one observation for each firm. I wanted to keep 25 observations for each firm.

[PeterClemmensen]

No. I assume that your data is sorted. Then the 25 first obs for each firm (stock in this case) are outputted

log:

1    data test;
2       do _N_ = 1 by 1 until (last.stock);
3          set sashelp.stocks;
4          by stock;
5          if _N_ <= 25 then output;
6       end;
7    run;

NOTE: There were 699 observations read from the data set SASHELP.STOCKS.
NOTE: The data set WORK.TEST has 75 observations and 8 variables.
NOTE: DATA statement used (Total process time):
      real time           0.13 seconds
      cpu time            0.04 seconds