Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: RE: Collinearity


From   jverkuilen <jverkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: Collinearity
Date   Wed, 10 Dec 2008 12:47:46 -0500

-findit collin-

It has VIF, condition number, etc. Often just looking at the correlation matrix or a matrix scatterplot helps, but no magic bullet exists for this problem. 

-----Original Message-----
From: "Steven Samuels" <sjhsamuels@earthlink.net>
To: statalist@hsphsun2.harvard.edu
Sent: 12/10/2008 10:27 AM
Subject: Re: st: RE: Collinearity

Regress x on the other covariates and you may detect which ones are  
collinear.  More than one may be implicated, especially if you have  
one too many dummy variables, as Martin suggested.

-Steve
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ismail  
> Ait Saadi
>
> I am running a Logit model with 12 variables, the results says that  
> variable
> x was dropped because of collinearity. is there any command that  
> can help
> identify the other variable with wich x is correlated, the reason I am
> asking for this is that x is an important variable and I prefer to  
> drop
> another variable and keep x. I remember doing this long time ago but I
> forgot the command.
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index