Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: negativ binomial operdispersion


From   Henrik Andersson <Henrik.Andersson@tft.lth.se>
To   statalist@hsphsun2.harvard.edu
Subject   st: negativ binomial operdispersion
Date   Fri, 14 Nov 2003 14:46:46 -0500

Hi,

I am using the nbreg command to run a negativ binomial regression. Using it I can choose between two different dispersion parameters, 

"dispersion(mean), which yields a model with dispersion equal to 1+alpha*exp(x_i*b +offset); i.e., the dispersion is a function of the expected mean: exp(x_i*b + offset). dispersion(constant) has dispersion equal to 1+delta; i.e., it is a constant for all observations." 

If alpha=0 or delta=0 I have a poisson model which would mean that dispersion=1.

The thing that I am interested in is what is ususlly referred to as the overdispersion parameter, which I need for a second stage estimation. My question is, therefore, is the estimated alpha or delta coefficient equal to this overdispersion parameter, i.e. without adding the constant 1? Or should I interpreted what STATA calls dispersion, 1+alpha*exp(x_i*b +offset) or 1+delta, as the overdispersion parameteter?

Best regards

Henrik Andersson


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index