Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Questions about gnpoisson
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: Questions about gnpoisson
Date
Thu, 20 Sep 2012 18:55:17 +0100
-gnpoisson- is a user-written command from SSC, as you explain.
I imagine that by phi is meant ln(phi) in the passage you cite.
The help file for -gnpoisson- gives references to two books, which no
doubt say more.
Please don't give partial name (date) references; this is an explicit
request in the FAQ.
Nick
On Thu, Sep 20, 2012 at 6:13 PM, <[email protected]> wrote:
> I'm using "gnpoisson", and having two questions about it.
>
> Firstly, according the help file of gnpoisson:
>
> “Positive values of phi adjust for Poisson overdispersion;
> negative values adjust for Poisson under-dispersion.”
>
> However, gnpoisson estimates ln(phi),
> and calculates the phi for the model output via exp(ln(phi)).
> Given this re-parameterisation, how can the phi be negative?
>
> Secondly, there are at least two kinds of generalised poisson
> model (Yang, Hardin, and Addy 2009), what kind of generalised
> poisson model is gnpoisson based on?
>
> (By the way, to install gnpoisson, just type:
> ssc install gnpoisson
*
* 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/