st: Negative Binomial Regression with traditional within unit fixedeffects?

[Michael E Roettger <meroett@bgsu.edu>]

Mike Roettger
Postdoctoral Fellow
National Center for Marriage Research
Bowling Green State University

Address:                                                                                                Phone:  419-372-6320
108 Bromfield Hall                                                                                  Fax:      419-372-3179
Bowling Green State University
Bowling Green, OH 43402

_________________________________________________________________

The more laws and order are made prominent,
The more thieves and robbers there will be.
   - Lao-tzu, Tao Te Ching
________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] On Behalf Of Michael E Roettger [meroett@bgsu.edu]
Sent: Tuesday, September 30, 2008 18:15
To: statalist@hsphsun2.harvard.edu
Subject: st: Thread-Index: AQHJI0ot1LurHl6hOky9rhtKsPbFYw==

Hi!

I have a dependent variable in panel data for individuals that fits a negative binomial distribution.  With an add on or extension, is it possible to run a negative binomial regression with fixed effects at the individual level?  My understanding is that

xtnbreg  Y X, fe

creates fe for distribution and not the xb variables for the regression equation y=XB+e.  I would like to generate individual level fixed effects for the XB variables, not the distribution.  Any advice for how to run a NB reg with fixed effects at the individual level  for the XB variables would be greatly appreciated!

Thanks,

Mike Roettger,
Bowling Green State University


*
*   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/