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: xtreg, fe and xtlogit, fe


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: xtreg, fe and xtlogit, fe
Date   Fri, 14 Oct 2011 09:22:19 +0200

On Thu, Oct 13, 2011 at 10:36 PM, natasha agarwal wrote:
> I did the xtreg model without the clustering of standard errors and it
> still gives me the same result.

That should be no surprise, as robust and clustered standard errors
only change the standard errors and leave the point estimates
unchanged, see [U] 20.16 (in the Stata 11 Users manual).

> But I guess you are right different
> models with different assumptions can altogether yield different
> results.

As I said before it is a very specific assumption: linear probability
models compare groups by looking differences in probabilities while
logistic regression models compare groups by looking at ratios of
odds. A ratio is a relative effect, while a difference is an absolute
effect. Both can be meaningful, but they can easily lead to apparently
different conclussions. Notice that this difference only apparent, you
can and should find out how these two findings fit together. Looking
at many graphs of predicted probabilities and odds, like Nick
suggested, is often helpful.

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/


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