用genmode过程做LOGISTIC回归时怎么出OR值啊？

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

用genmode过程做LOGISTIC回归时在哪个位置写什么命令才能出OR值啊？

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1 回帖

关键词：logistic回归 logistic logisti ogistic logist

wait for the professional

proc genmod data=resp descend;
   class id treatment(ref="P") center(ref="1") sex(ref="M")
      baseline(ref="0") / param=ref;
   model outcome=treatment center sex age baseline / dist=bin;
   repeated  subject=id(center) / logor=fullclust;
run;

LOGOR=log-odds-ratio-structure-keyword specifies the regression structure of the log odds ratio used to model the association of the responses from subjects for binary data. The response syntax must be of the single variable type, the distribution must be binomial, and the data must be binary.

   data resp;
      keep id active center center2 female age baseline visit outcome;
      input center id treatmnt $ sex $ age baseline visit1-visit4;
      active=(treatmnt='A');
      center2=(center=2);
      female=(sex='F');
      visit=1;  outcome=visit1;  output;
      visit=2;  outcome=visit2;  output;
      visit=3;  outcome=visit3;  output;
      visit=4;  outcome=visit4;  output;
      cards;
   1  1 P M 46 0 0 0 0 0
   1  2 P M 28 0 0 0 0 0
   1  3 A M 23 1 1 1 1 1
   1  4 P M 44 1 1 1 1 0
   1  5 P F 13 1 1 1 1 1
   1  6 A M 34 0 0 0 0 0
   1  7 P M 43 0 1 0 1 1
   1  8 A M 28 0 0 0 0 0
   1  9 A M 31 1 1 1 1 1
   1 10 P M 37 1 0 1 1 0
   1 11 A M 30 1 1 1 1 1
   1 12 A M 14 0 1 1 1 0
   1 13 P M 23 1 1 0 0 0
   1 14 P M 30 0 0 0 0 0
   1 15 P M 20 1 1 1 1 1
   1 16 A M 22 0 0 0 0 1
   1 17 P M 25 0 0 0 0 0
   1 18 A F 47 0 0 1 1 1
   1 19 P F 31 0 0 0 0 0
   1 20 A M 20 1 1 0 1 0
   1 21 A M 26 0 1 0 1 0
   1 22 A M 46 1 1 1 1 1
   1 23 A M 32 1 1 1 1 1
   1 24 A M 48 0 1 0 0 0
   1 25 P F 35 0 0 0 0 0
   1 26 A M 26 0 0 0 0 0
   1 27 P M 23 1 1 0 1 1
   1 28 P F 36 0 1 1 0 0
   1 29 P M 19 0 1 1 0 0
   1 30 A M 28 0 0 0 0 0
   1 31 P M 37 0 0 0 0 0
   1 32 A M 23 0 1 1 1 1
   1 33 A M 30 1 1 1 1 0
   1 34 P M 15 0 0 1 1 0
   1 35 A M 26 0 0 0 1 0
   1 36 P F 45 0 0 0 0 0
   1 37 A M 31 0 0 1 0 0
   1 38 A M 50 0 0 0 0 0
   1 39 P M 28 0 0 0 0 0
   1 40 P M 26 0 0 0 0 0
   1 41 P M 14 0 0 0 0 1
   1 42 A M 31 0 0 1 0 0
   1 43 P M 13 1 1 1 1 1
   1 44 P M 27 0 0 0 0 0
   1 45 P M 26 0 1 0 1 1
   1 46 P M 49 0 0 0 0 0
   1 47 P M 63 0 0 0 0 0
   1 48 A M 57 1 1 1 1 1
   1 49 P M 27 1 1 1 1 1
   1 50 A M 22 0 0 1 1 1
   1 51 A M 15 0 0 1 1 1
   1 52 P M 43 0 0 0 1 0
   1 53 A F 32 0 0 0 1 0
   1 54 A M 11 1 1 1 1 0
   1 55 P M 24 1 1 1 1 1
   1 56 A M 25 0 1 1 0 1
   2  1 P F 39 0 0 0 0 0
   2  2 A M 25 0 0 1 1 1
   2  3 A M 58 1 1 1 1 1
   2  4 P F 51 1 1 0 1 1
   2  5 P F 32 1 0 0 1 1
   2  6 P M 45 1 1 0 0 0
   2  7 P F 44 1 1 1 1 1
   2  8 P F 48 0 0 0 0 0
   2  9 A M 26 0 1 1 1 1
   2 10 A M 14 0 1 1 1 1
   2 11 P F 48 0 0 0 0 0
   2 12 A M 13 1 1 1 1 1
   2 13 P M 20 0 1 1 1 1
   2 14 A M 37 1 1 0 0 1
   2 15 A M 25 1 1 1 1 1
   2 16 A M 20 0 0 0 0 0
   2 17 P F 58 0 1 0 0 0
   2 18 P M 38 1 1 0 0 0
   2 19 A M 55 1 1 1 1 1
   2 20 A M 24 1 1 1 1 1
   2 21 P F 36 1 1 0 0 1
   2 22 P M 36 0 1 1 1 1
   2 23 A F 60 1 1 1 1 1
   2 24 P M 15 1 0 0 1 1
   2 25 A M 25 1 1 1 1 0
   2 26 A M 35 1 1 1 1 1
   2 27 A M 19 1 1 0 1 1
   2 28 P F 31 1 1 1 1 1
   2 29 A M 21 1 1 1 1 1
   2 30 A F 37 0 1 1 1 1
   2 31 P M 52 0 1 1 1 1
   2 32 A M 55 0 0 1 1 0
   2 33 P M 19 1 0 0 1 1
   2 34 P M 20 1 0 1 1 1
   2 35 P M 42 1 0 0 0 0
   2 36 A M 41 1 1 1 1 1
   2 37 A M 52 0 0 0 0 0
   2 38 P F 47 0 1 1 0 1
   2 39 P M 11 1 1 1 1 1
   2 40 P M 14 0 0 0 1 0
   2 41 P M 15 1 1 1 1 1
   2 42 P M 66 1 1 1 1 1
   2 43 A M 34 0 1 1 0 1
   2 44 P M 43 0 0 0 0 0
   2 45 P M 33 1 1 1 0 1
   2 46 P M 48 1 1 0 0 0
   2 47 A M 20 0 1 1 1 1
   2 48 P F 39 1 0 1 0 0
   2 49 A M 28 0 1 0 0 0
   2 50 P F 38 0 0 0 0 0
   2 51 A M 43 1 1 1 1 0
   2 52 A F 39 0 1 1 1 1
   2 53 A M 68 0 1 1 1 1
   2 54 A F 63 1 1 1 1 1
   2 55 A M 31 1 1 1 1 1
    ;

proc genmod data=resp descend;
class id center;
model outcome=center2 active female age baseline / dist=bin;
repeated  subject=id(center) / logor=fullclust;
run;

logor= gives the Alternative Logistic Regression Algorithm for repeated measurements, which is irrelevant to the question.
In genmod, there is no way to directly give OR in parameter estimates output, as far as I know. what we normally do is taking the Parameter Estimates first and then doing the exponential calculation.
another way to work around the problem is through using statement ESTIMATE/EXP syntax but it needs you to write out the correct estimate formula, which often intimidates some beginners.
JingJu

jingju11 发表于 2012-11-12 01:24
logor= gives the Alternative Logistic Regression Algorithm for repeated measurements, which is irrel ...

这个回答是靠谱的。
proc genmod data=fmpr descend;
class fno isb4(ref="3") /param=ref;
model glucose=isb4 /dist=bin;
repeated subject=fno/corr=un;
estimate "OR：0 vs 3" isb4 1 0 0 -1/exp;/**第一组与第三组（对照）相比的OR**/;
run;