logistic回归中，当数据中含加权值而反应变量是一个二分变量,应使用MODEL Y/N=X1 X2......的方式来界定模型而非 MODEL Y=X1 X2... 的 方式 。也就是找出active/k与x y 之间的关系。
data temp;
input active x y z k;
cards;
1 2 45 123 1200
0 10 45 222 2400
1 14 56 . 1400
1 22 47 200 1500
0 9 35 . 1700
1 12 26 455 2000
;
run;
proc logistic data=temp descending;
model active/k=x y ;
output out=temp1 pred=pred;
weight  z;
run;

2# 坐看云起时


的确是这样，谢谢阁下的解答，谢谢！

2# 坐看云起时

希望赐教：

proc logistic data=temp descending;
model active/k=x y ;
output out=temp1 pred=pred;
weight  z;
run;
我对model active/k=x y ;的理解是 active =事件发生的次数，k=实验的次数；

可是原始的程序好像是active是0 1 反应变量；另外原始程序里说他运用的是少量数据，其实按照active/k 的写法，数据并不小，只是事件发生的次数少而已；
另外楼主认为程序中断运行是因为权重的关系。缘何判断呢？有没有可能过大的权重以及过少的观察数造成算法出错呢？比如说改写程序：

proc logistic data=temp descending;
freq K;
model active=x y ;
output out=temp1 pred=pred;
weight  z;
run;
此处增多的观察数保证程序运行无误。

MODEL Statement
MODEL events/trials= < effects > < / options >;
MODEL variable < (variable_options) >= < effects > < / options >;

Two forms of the MODEL statement can be specified. The first form, referred to as single-trial syntax, is applicable to binary, ordinal, and nominal response data. The second form, referred to as events/trials syntax, is restricted to the case of binary response data. The single-trial syntax is used when each observation in the DATA= data set contains information on only a single trial, for instance, a single subject in an experiment. When each observation contains information on multiple binary-response trials, such as the counts of the number of subjects observed and the number responding, then events/trials syntax can be used.

In the events/trials syntax, you specify two variables that contain count data for a binomial experiment. These two variables are separated by a slash. The value of the first variable, events, is the number of positive responses (or events). The value of the second variable, trials, is the number of trials. The values of both events and (trials-events) must be nonnegative and the value of trials must be positive for the response to be valid.

In the single-trial syntax, you specify one variable (on the left side of the equal sign) as the response variable. This variable can be character or numeric. Options specific to the response variable can be specified immediately after the response variable with a pair of parentheses around them.

For both forms of the MODEL statement, explanatory effects follow the equal sign. Variables can be either continuous or classification variables. Classification variables can be character or numeric, and they must be declared in the CLASS statement. When an effect is a classification variable, the procedure enters a set of coded columns into the design matrix instead of directly entering a single column containing the values of the variable.

你也可以查阅“《世界统计与分析全才SAS系统应用指南”p387页关于于逻辑回归中对权数设定时，model的要求。

5# 坐看云起时

感谢你的回复。只是看大段的英文解释让人头痛。我没有那本书，暂时没法查阅。
举例：

p ro c    l o g i s t i c d a t a = t o d a y ;
  model   d i s e a s e = f ema l e ;
  we i g h t     we i g h t ;
run；
从此处而来 http://www.biostat.umn.edu/~melanie/PH7402/2009/NOTE/proc%20logistic%20traps.pdf

我的理解是，model的写法，要采用第一或者第二种方法， 与是否含有weight statement无关。model的写法只是取决于data 的形式。如果是weight发生错误，不应该用改变model的写法来解决问题。K如果在等式的右边，那就是个自变量。如果是 */K= ， 在等式的左边，那它就是实验的次数。

如果按照 model active/k=x y；我的理解还是：very few events occurred on a large numer of observations. 去掉missing，total number of responses（events）=3, and total number of trials=1200+2400+1500+2000=7100. 这也是我为什么说这是一个很大的dataset。既然data很大，此时用weight，没有运算错误也是可以理解的。

如果按照 model active=x y z。此时total number of responses（events）=3, and total number of observations=4.;这是一个很小的data。

所以，我的理解还是，你和原始程序用的是不同的数据。

input active x y z k;
cards;
1 2 45 123 1200
0 10 45 222 2400
1 14 56 . 1400
1 22 47 200 1500
0 9 35 . 1700
1 12 26 455 2000
;
run;