SAS Logistic 输出结果中Percent Tied什么意思呢

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

请教各位高手：SAS Logistic 输出结果中Percent Tied什么意思呢  

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：logistic logisti ogistic percent logist

顶起.JingJu

Shamelessly copied from SAS documentation,
"A pair of observations with different observed responses is said to be concordant if the observation with the lower ordered response value has a lower predicted mean score than the observation with the higher ordered response value. If the observation with the lower ordered response value has a higher predicted mean score than the observation with the higher ordered response value, then the pair is discordant. If the pair is neither concordant nor discordant, it is a tie. Enumeration of the total numbers of concordant and discordant pairs is carried out by categorizing the predicted mean score into intervals of length  and accumulating the corresponding frequencies of observations. Note that the length of these intervals can be modified by specification of the BINWIDTH= option in the MODEL statement. "
So, it's easy to calcuate the percentages by yourself. Just keep in mind a pair ties if the predicted probability lies in the same BIN.

1. data vaso;
   
2.    length Response $12;
   
3.    input Volume Rate Response @@;
   
4.    LogVolume=log(Volume);
   
5.    LogRate=log(Rate);
   
6.    datalines;
   
7. 3.70  0.825  constrict       3.50  1.09   constrict
   
8. 1.25  2.50   constrict       0.75  1.50   constrict
   
9. 0.80  3.20   constrict       0.70  3.50   constrict
   
10. 0.60  0.75   no_constrict    1.10  1.70   no_constrict
    
11. 0.90  0.75   no_constrict    0.90  0.45   no_constrict
    
12. 0.80  0.57   no_constrict    0.55  2.75   no_constrict
    
13. 0.60  3.00   no_constrict    1.40  2.33   constrict
    
14. 0.75  3.75   constrict       2.30  1.64   constrict
    
15. 3.20  1.60   constrict       0.85  1.415  constrict
    
16. 1.70  1.06   no_constrict    1.80  1.80   constrict
    
17. 0.40  2.00   no_constrict    0.95  1.36   no_constrict
    
18. 1.35  1.35   no_constrict    1.50  1.36   no_constrict
    
19. 1.60  1.78   constrict       0.60  1.50   no_constrict
    
20. 1.80  1.50   constrict       0.95  1.90   no_constrict
    
21. 1.90  0.95   constrict       1.60  0.40   no_constrict
    
22. 2.70  0.75   constrict       2.35  0.03   no_constrict
    
23. 1.10  1.83   no_constrict    1.10  2.20   constrict
    
24. 1.20  2.00   constrict       0.80  3.33   constrict
    
25. 0.95  1.90   no_constrict    0.75  1.90   no_constrict
    
26. 1.30  1.625  constrict
    
27. ;
    
28. proc logistic data=vaso;
    
29.    model Response=LogRate LogVolume/binwidth=0.1;
    
30.    output out=xxx predicted=phat;
    
31. run;
    
32. proc sql;
    
33. create table pairs as
    
34.         select ceil(10*a.phat) as rank0, ceil(10*b.phat) as rank1
    
35.         from xxx a, xxx b
    
36.         where a.response like 'n%' and b.response like 'c%';
    
37. quit;
    
38. 
    
39. data hell;
    
40.         set pairs nobs=denominator end=last;
    
41.         if rank0=rank1 then tied+1;
    
42.         else if rank0<rank1 then concordant+1;
    
43.         else discordant + 1;
    
44.         if last then do;
    
45.                 p_tied = tied/denominator;
    
46.                 p_concordant = concordant/denominator;
    
47.                 p_discordant = discordant/denominator;
    
48.                 output;
    
49.         end;
    
50.         keep tied concordant discordant p_:;
    
51.         label p_tied = '% tied' p_concordant='% concordant'
    
52.                 p_discordant='% discordant';
    
53.         format p_: percent7.1;
    
54. proc print label;

复制代码

Example copythefted from
http://support.sas.com/documenta ... ogistic_sect064.htm

thank you very much!!!