Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Need Help - Questions about Count Data Panel Estimation


From   "Jerry Margolis" <[email protected]>
To   [email protected]
Subject   st: Need Help - Questions about Count Data Panel Estimation
Date   Wed, 6 Jun 2007 13:38:36 +1000

Hi fellow Stata users

I'm doing count panel estimation, and I am puzzled by the following:

About xtnbreg:

1. Why is it that the xtnbreg command does not put the fixed effects
and random effects in the X'T component, in the same way that panel
estimators are constructed like, the xtpoisson?


2.  Why is there no overdispersion test for xtnbreg, unlike its
cross-section counterpart nbreg?

About Spost command written by Scott long and Frees:

I understand that this command allows us to create a graph to compare
model fit between the Poisson and Negative binomial for cross section
data. That is, how well the predicted probability of counts given by
the Poisson and Negative binomial match the actual, observed
proportion of counts. Unfortunately, this command does not work with
xtpoisson and xtnbreg regressions. I was wondering if there is a
similar user-friendly way to compare model fit for panel data, or if
there is actually no need to do so, and the cross section results
using Spost carry over to pandel data.

Thanks,


Jerry
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index