非常感谢老师的回复，之前为什么我的程序中出来三个结果不一样，主要是因为选择的class变量分层太多，所以模型无法很好收敛。老师生成的新的class变量，保证了三个模型都能很好收敛。所以也就是说，只有在按照class变量分层后，各层样本量相对充足的情况下，保证模型能够充分收敛，三者才能取得理论上一致的结果。
另外，上述讨论的是三者之间的一致性，我已经相对比较明白，但是如果genmod只是能构建logisitic回归相同的模型，那genmod与logisitic回归在处理二分类结局变量资料上又有什么区别呢，也就是genmod相对于logisitic回归在处理二分类结局变量资料上又有什么优势呢。谢谢老师指点。

Agree.

The more relevant reason for non-convergence here is because of data separation. This kind of data or model specification will cause computation matrix not stable or singular. for example, there is no inverse matrix available. you may observe this from an extremely large standard errors for estimated parameters. similar to 1/0.

Genmod covers many distributions from exponential family. Of note, Genmod provides GEE method that is for longitudinal data. Of course, PROC GEE is available at SAS/STAT 12.3, which may give more options.
JingJu

谢谢，线性可分确实是logisitc回归使用的时候应该注意的问题。如果数据确实存在class变量导致的不收敛以及线性可分的问题，采用上述三种方法来分析二分类结局变量资料的影响因素时，是否都不合适？还是说除了简单地的变量变换外，还有其他的方法来处理此类问题？

请问你这个平滑图是怎么画出来的呢？

请问会poisson回归模型不

谢谢分享

感谢楼主分享，这里的样条 和rcs区别大吗？ 非线性部分，作图的Y轴可以看做是B吗？回归系数？

请问您知道为什么要在构建gam模型的时候，给每个变量加上自由度DF吗？

在SAS 论坛上看到如下答复，一定程度上可以回应楼主问题：
The results are consistent with each other.  Both are correct.  However, they have been obtained using different expansions of the categorical variables.  The GENMOD procedure employs an over parameterized model in which a set of k binary variables are produced when the number of levels of a categorical variable is k.  SAS refers to this as the GLM parameterization.  By default, the LOGISTIC procedure employs a model with k-1 variables in the design matrix.  Moreover, the k-1 variables are not binary, but can take on one of three values: -1, 0, or 1. This sort of parameterization is referred to as effect coding.

For variable X1, columns of the design matrix given GLM coding and effect coding are as follows:

             GLM coding

   X1    X1_1    X1_2    X1_3

    a     1       0       0

    b     0       1       0

    c     0       0       1

        Effect coding

   X1    X1_1    X1_2

    a     1       0

    b     0       1

    c    -1      -1

It can be shown that these two parameterizations will yield the same predicted response values.  But the parameters do have to be interpreted differently.

If you prefer the GLM parameterization (a lot of people do), you can request that parameterization in the LOGISTIC procedure.  All you have to do is change your class statement to:

class x1 x1 / param=glm;

来源：
https://communities.sas.com/t5/Statistical-Procedures/PROC-LOGISTIC-vs-PROC-GENMOD-different-results/m-p/27600#M1039

谢谢分享！