參考 http://www.nesug.org/proceedings/nesug00/cc/cc4004.pdf

其中提到"Because no observations are read in or written out during processing, for an existing dataset, the MODIFY command is the best way to add a label, rename a variable, or change a format or informat."

modify比較適合作為修改資料或是變數屬性
若是要更動到整個資料檔，應該用set會比較好

希望有回答到你的問題^^

没人么？自己顶一下

用你寫的程式跑了SAS是OK的...
樣本數最後剩下50個
我的SAS版本是9.2

NOTE: SAS initialization used:
      real time           1.26 seconds
      cpu time            0.96 seconds

1    data mst;
2      do i=1 to 100;
3       output;
4      end;
5    run;

NOTE: The data set WORK.MST has 100 observations and 1 variables.
NOTE: DATA statement used (Total process time):
      real time           0.01 seconds
      cpu time            0.00 seconds


6
7    data mst;
8      modify mst;
9      if i le 50 then remove;
10   run;

NOTE: There were 100 observations read from the data set WORK.MST.
NOTE: The data set WORK.MST has been updated.  There were 0 observations rewritten, 0
      observations added and 50 observations deleted.
NOTE: DATA statement used (Total process time):
      real time           0.01 seconds
      cpu time            0.00 seconds

对，样本数最后的确剩下50个，但打开数据集，观测序号是从51开始的，而非从1开始，并且数据集大小没有减少，我想问的是为什么数据集大小不变

若是把修改過後的data另存成新的檔(temp)
就可以看出觀測值僅剩50個
觀測序號也是從1開始
但就不是從修改原始檔的做法了...

data mst;
  do i=1 to 100;
   output;
  end;
run;

data mst;
  modify mst;
  if i le 50 then remove;
run;

data temp;
        set mst;
run;