[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re: Poolability tests and count data
Both xtpois and xtnbreg provide a likelihood ratio test of the panel estimator
with the pooled estimator.
----- Original Message -----
Sent: Thursday, October 17, 2002 9:17 AM
Subject: st: Poolability tests and count data
> i am going to work with a panel (300 firms, information from 18 years each)
and want to know if i "can" pool the data or if i have to use panel-methods. The
well known tests like the LR- or Breusch-Pagan-Test are always used with
non-discrete data but not with count data. My dataset contains count data and i
want to know if i should use the NEGBIN-modell or the random or fixed
effect-panel modell with count data. Which test do i have to make and how can i
perform it with STATA?
> Yours sincerly,
* For searches and help try: