不要不理我啦~~~肿么办呢？

可以重复取样，自己看帮助

用permutation试试看~~~

Here is a simple one with data step.

data sample;
array dates [10] ('1jan2001'd,'1may2009'd,'1may2010'd,'1apr1906'd,'1feb2010'd,
                  '1aug2001'd,'1may2012'd,'1may1935'd,'1apr1840'd,'1feb1999'd);
seed=534;
do idnub=1 to 2400;
   n=rantbl(seed,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1,0.1);
   date=dates[n];
   output;
end;
format date yymmdd10.;
run;

proc freq data=sample;
table date;
run;

找了半天。。还没找到，那我再找找吧

data a;
input id $ realday spending;
datalines;
10000501 27 27.6
10000501 187 3.9
10000501 201 9.1
10000502 218 3.5
10000501 249 3.6
10000501 270 3.5
10000501 274 8.3
10000501 292 35.7
10000502 322 2.3
10000501 333 7.9
10000501 343 36
10000502 434 4.4
10000501 476 7.9
10000502 486 5
10000501 504 10.2
10000501 508 3.6
;
run;


proc surveyselect data=a
      method=urs n=100 outhits
      out=Sample;
run;

我用了urs，然后的确能抽出100个数据，但是他的排序是根据数据集a来的，有哪种显示方法可以是随机显示，而不是排好序的？

这个方法的确可以实现。。。。。谢谢啦！！！