st: RE: RE: Using xtprobit pa, xtprobit re or xtlogit fe ?

["Verpoorten, Marijke" <Marijke.Verpoorten@econ.kuleuven.be>]

Xtlogit, fe drops all observations for which the dependent variable is
always 0 or always 1. This is because of the conditional maximum
likelihood estimation procedure that uses the average outcome as a
sufficient statistic, (cf. Verbeek, 2000 or any other handbook that
treats binary panel data)

Marijke

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Scott
Merryman
Sent: zaterdag 9 december 2006 18:11
To: statalist@hsphsun2.harvard.edu
Subject: st: RE: Using xtprobit pa, xtprobit re or xtlogit fe ?

Was there not a message at the top of the output following -xtlogit,fe-
stating why they were dropped?  Something like:

note: 108 groups (453 obs) dropped due to all positive or
      all negative outcomes.

Scott


> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of Claude Francoeur
> Sent: Saturday, December 09, 2006 9:30 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Using xtprobit pa, xtprobit re or xtlogit fe ?
> 
> Hello,
> 
> My data is organized in a panel (100 firms over 5 years) with a binary
> dependent variable.
> 
> I get interesting results, using xtprobit, pa or re (N = 581 and no of
> groups = 135). Here are the commands:
> 
> xtprobit pw boardcomp compensation shrights disclosure tech mv endet
cot
> volat march conc, pa i(i)
> 
> xtprobit pw boardcomp compensation shrights disclosure tech mv endet
cot
> volat march conc, re i(i)
> 
> 
> Using xtlogit and controlling for fixed effects, all coefficients
become
> non
> significant (N = 128 and no of groups = 27)
> 
> xtlogit pw boardcomp compensation shrights disclosure tech mv endet
cot
> volat march conc, fe i(i)
> 
> Why does xtlogit eliminate so many observations and all significance?
Is
> xtlogit appropriate ?
> 
> Thanks,
> 
> Claude
> 


*
*   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/

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm


*
*   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/