批量计算fisher概率值

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

有一张表，有3000多行，每行有一个a、b、c、d值，如图所示，a、b、c、d表示四格表的数字，我想对每一行的数据都计算列联系数C和fisher概率p值，并和原始数据一起输出到一张表中，每一行要求有原始的数字及计算出的联系数C和fisher概率p值，如图所示，请问怎么完成啊？

a	b
c	d

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Fisher Fish SHE fis 如图所示 fisher 计算

I don't really know what C stands for, but for Fishers' P: jingju
If you want to self-define a function to compute Fishers' P value, you can refer to the blog. But you may notice that computing fishers' is involving lots of calculations, thus might-be over-floating during the computation. that is one reason why calculating fishers' in more than 2 by 2 is time-consuming.

http://blog.sina.com.cn/s/blog_a3a92636010134mn.html

1. data had;
   
2.    a= 232; b =13; c=59; d=1153; output;
   
3.    a= 232; b =59; c=13; d=1153; output;
   
4.    a=1313; b =49; c= 1; d= 94; output;
   
5. run;
   
6. data have;
   
7.    set had;
   
8.    array n[2,2] a b c d;
   
9.    v_by ++1;
   
10.    do i =1 to dim1(n);
    
11.       do j =1 to dim2(n);
    
12.          count =n[i,j];output;
    
13.       end;
    
14.    end;
    
15.    keep v_by i j count;
    
16. run;
    
17. ods output fishersexact =FishersExact(where =(name1 ="XP2_FISH"));
    
18. proc freq data =have order=data;
    
19.    by v_by;
    
20.    tables i*j / chisq;
    
21.    weight Count;
    
22. run;
    
23. data have2;
    
24.    merge had FishersExact(keep =nValue1 rename =(nValue1 = p));
    
25. run;

复制代码

学习