Remove intercept with "regress"

3 次查看（过去 30 天）
Dan Pron
Dan Pron 2015-4-17
回答： Dan Pron 2015-4-17
Hello everyone!
I would like to run a multiple linear regression, but dont want any intercept in the output (i want to remove it, because i'm facing a multicolinearity issue). Can someone help me? Here is a piece of my code :
[b17, bint17,~,~,stats17] = regress(TrimROA, [ones(length(TrimROA),1) TrimOPEXP_LP TrimBPSM TrimWO TrimCPB TrimDEBT_EQU],0.1)
Thanks a lot!

请先登录，再回答此问题。

回答（2 个）

Chris McComb
Chris McComb 2015-4-17
I think that all you need to do is remove the first column from your second argument (i.e. the column of 1's).
[b17, bint17,~,~,stats17] = regress(TrimROA, [TrimOPEXP_LP TrimBPSM TrimWO TrimCPB TrimDEBT_EQU],0.1)
Dan Pron
Dan Pron 2015-4-17
Thanks, but i still have an issue^^. I removed the column of ones :
ones(length(TrimROA),1)
But i got this error message now :
Warning: R-square and the F statistic are not well-defined unless X has a column of ones.
Type "help regress" for more information.
> In regress at 163
This is a problem, because i need to have a reliable R-square for my work. Any idea?

类别

在 Help Center 和 File Exchange 中查找有关 Linear Regression 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by