0、1变量可视作虚拟变量，不正好省事了吗？

proc logistic data=XX descending;

model y=X1 X2;

run;

after you get the coefficients then you can get the P for y=1

data own;
input type x1-x3;
cards;
0 11 0
1 13 0
1 14 0
0 12 0
1 18 1
1 20 1
0 9 0
1 13 1
0 12 0
1 14 0
1 13 1
0 15 0
1 11 1
;
proc logistic;
model type=x1-x3/selection=stepwise sle=0.3 sls=0.3;
output out=result p=predi xbeta=XI;run;
proc print data=result;run;
有大侠能帮小弟改下吗？

不可取。

这个不能用虚拟变量来代替，代替了的话，就没有分析的意义了

Faint , you have only 2 predictors , no need to do selection