Re: st: nbreg with fixed effect vs xtnbreg,fe

[Richard Williams <richardwilliams.ndu@gmail.com>]

At 08:52 PM 2/7/2012, Shikha Sinha wrote:
> Hi all,
>
> I emailed my query to tech support at Stata corp and below is the response;
>
>
> Typically for a fixed effects negative binomial model, you would want to use
> the -xtnbreg, fe- command.   -xtnbreg, fe- is fitting a conditional fixed
> effects model.  When you include panel dummies in -nbreg- command, you are
> fitting an unconditional fixed effects model.  For nonlinear models such as
> the negative binomial model, the unconditional fixed effects estimator
> produces inconsistent estimates.  This is caused by the incidental parameters
> problem.  See the following references for theoretical aspects on the
> incidental parameters problem:
>
>                Greene, William H. "Econometric Analysis". Prentice Hall.
>                Seventh Edition, page 413.
>
>                Baltagi, Badi "Econometric Analysis of Panel Data".
>                        4th. Edition. John Wiley and Sons LTD.
>                        Section 11.1 (pages 237-8).

Again, unless I am missing something, Allison disagrees. On p. 64 of 
his book, he says "Using Monte Carlo simulations, Allison and 
Waterman found that the unconditional negative binomial estimator did 
not show any substantial bias from incidental parameters." He does 
add that the unconditional confidence intervals tended to be too 
small, but he shows how to easily correct for that. He then gives 
examples that he claims show that -xtnbreg, fe- is not a true fixed 
effects method.

Again, Allison's book is "Fixed effects regression models", Sage, 
2009. His 2002 paper that his book draws from is

Allison, Paul D. and Richard Waterman (2002) "Fixed effects negative 
binomial regression models." In Ross M. Stolzenberg (ed.), 
Sociological Methodology 2002. Oxford: Basil Blackwell.

Again, I might be screwing up Allison's argument. But if not, I'd 
like to see him and StataCorp go 15 rounds on this. ;-)


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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/