Dear,
I am afraid that -posgof- is only available after a -poisson- analysis,
and not in the case of panel data -xtpois- analysis.
Kind regards,
Ad van den Oord
-----Original Message-----
From: Laurel.Copeland@med.va.gov [mailto:Laurel.Copeland@med.va.gov]
Sent: Monday, March 22, 2004 2:11 PM
To: statalist@hsphsun2.harvard.edu
Cc: Oord, A. van den
Subject: RE: xtpois versus xtnbreg
I believe issuing -poisgof- after the -poisson- analysis will do it.
-----Original Message-----
From: Oord, A. van den [mailto:A.v.d.Oord@tm.tue.nl]
Sent: Monday, March 22, 2004 6:12 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: xtpois versus xtnbreg
Hi All,
I have a data set where I count the number of patents for a group of 113
companies per year over a period of 12 years. For my analysis, I
currently use a random effects poisson model for panel data (the command
'xtpois'). However, to be certain that I am using the correct model, I
wish to determine whether my dependent variable is overdispersed or not.
In the case of overdispersion I have to use the random effects negative
binomial model for panel data (the command 'xtnbreg'). How can I
determine whether my dependent variable is overdispersed or not?
Thanks in advance,
Ad van den Oord
*
* 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/
*
* 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/
