Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Xtivreg2: Dropping dummy variables due to Collinearity


From   Shonda Naples <liminalzest@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Xtivreg2: Dropping dummy variables due to Collinearity
Date   Tue, 14 Jun 2011 10:56:35 -0400

An alternative to what Maarten suggests, you could drop the intercept
term and retain all the dummies.

On Tue, Jun 14, 2011 at 9:52 AM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Tue, Jun 14, 2011 at 3:27 PM, vikramfinavker wrote:
>> I am using xtivreg2 and the regression is dropping my dummy variables for
>> year due to collinerarity. I have 10 year dummy:
>
>> In normal regression (reg) it omits yr99 and yr00 saying there is
>> collinearity but in xtivreg2 it drops yr99 and yr09 saying they are
>> collinear.
>>
>> Could someone please explain me whats wrong here? Is it my data?
>
> When a set of variables are perfectly colinear Stata cannot
> distinguish between them, so it is pretty much random which variables
> it will exclude. For predictions it won't matter (which is why Stata
> cannot distinguish between them), but for interpretation of your
> results it matters a lot. A computer cannot know how you want to
> interpret your results. So if you don't tell it, the computer must
> ignore that consideration. So, you must tell Stata how you want to
> interpret your year dummies. Dummies are (usually, but there are
> exceptions) comparisons of groups in your case years, so you must tell
> Stata which year you want to compare with which other year. You
> typically do so by leaving out one year dummy, which is the reference
> category (but there are other ways of doing it). All remaining year
> dummies are comparison with that reference category.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> *   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   |   Site index