|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: xtlogit and constant term, thanks
Thanks Nicola,
I'd like to add something I found: the random effects specification (default for xtlogit in Stata)
does consider the constant, though it is dropped by the fixed effects specification (see example
from stalist http://www.stata.com/statalist/archive/2004-01/msg00669.html )
Juan Julio
--- [email protected] wrote:
> I had a similar problem and the professor of statistics at my PhD program said that omitting the
> constant radically changes your model. You should never omit it because if an intercept "really"
> exists, it is not wise to ask Stata to put it = 0; if the intercept is "really" zero, Stata will
> find it.
> Nicola
>
> At 02.33 10/01/2007 -0500, Juan Julio Gutierrez <[email protected]> wrote:
> >Deal all
> >
> >I am regressing the same model w/ and w/out the intercept using "xtlogit, re" in an unbalanced
> >panel data. However the odds' signs change if the constant is dropped.
> >
> >Could you help me explain this? (I am copying both outputs)
> >I appreciate you help
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
