SAS中reg回归变量系数估计为0

相似文件 换一批

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

求助大家帮忙看一下，cb该如何解释，它的系数估计为0...
cb=change*bs交叉项

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0 回帖

关键词：变量系数 REG change 求助大家 Hang change 如何

夏之鹭江 发表于 2015-2-25 18:16
求助大家帮忙看一下，cb该如何解释，它的系数估计为0...
cb=change*bs交叉项

collinear

偏回归系数为零，含义是在其他变量保持不变的情况下cb的变化不会引起因变量的变化。

Yeah.
As mentioned above, but this is the strongest collinearity. Somehow, cb can be expressed as
cb = c1+c2*bs, where both c1 and c2 are constants. That also causes biased estiamtes for both intercept and bs.

I have simulated the cases like what you have shown:see my example
JingJu

I have simulated the cases you have mentioned.

see my example

JingJu

jingju11 发表于 2015-2-26 08:38
I have simulated the cases you have mentioned.

see my example

京剧大师，请问下自由度B是啥情况？

liudeng2005 发表于 2015-2-26 09:21
京剧大师，请问下自由度B是啥情况？

DF=B represents that the corresponding estimate is biased. in this case, estimates of intercept and bs are both biased.
JingJu

jingju11 发表于 2015-2-26 08:38
I have simulated the cases you have mentioned.

see my example

谢谢京剧前辈