|
楼主: a125685446
|
1286
2
真的需要你的帮助 |
|
大专生 58%
-
|
回帖推荐You may use vif and collin on model statement for Variance Inflation Index and Condition Number, both are good indicators for collinearity. Here is an example. Hope it helps.
data t1;
do i = 1 to 100;
x=rannor(123);
x2=x+0.001*rannor(123);
err= rannor(123);
y=1+ x+x2 + err;
output;
end;
run;
proc reg data=t1;
model y=x x2/tol vif collin ;
run;
quit;
本帖被以下文库推荐
| ||
|
|
| ||
加好友,备注jltj京ICP备16021002号-2 京B2-20170662号
京公网安备 11010802022788号
论坛法律顾问:王进律师
知识产权保护声明
免责及隐私声明


