Statalist The Stata Listserver


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

RE: st: Zero Inflated Negative Binomial Model on Panel Data?


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Zero Inflated Negative Binomial Model on Panel Data?
Date   Tue, 18 Jul 2006 22:12:18 +0100

I guess the reasons no one answered your previous posting were
tbat 

1. No such command exists, although individuals might be 
wary of asserting that. 

2. It is very rare that undocumented options exist to do 
important things. Thus if the options -re- or -fe- were 
allowed with -zinb- there is no obvious reason why 
StataCorp would want to hide such important functionality. 
But if the manual doesn't document it and attempts to use it 
fail you are near a proof of non-existence for that command. 
(I don't think a panel -zinb- would be called -zinb-, given Stata's naming 
practices, which is another hint.)  

Nick 
n.j.cox@durham.ac.uk 

Sandip Basu
 
> Thanks, bw, for the effort in looking up the reference. You 
> are probably right about the panel data option not being 
> available in zero inflated models. Anyway, I should heve been 
> more specific...... the program did not recognize the "re" or 
> "fe" command together with "zinb". Without "re" or "fe" there 
> was no mention of the number of groups in the output. Hence, 
> I was hoping that someone had a suggestion for an alternative 
> or additional command that would recognize the group 
> memebership in the panel data.
 
On Tue, 18 Jul 2006, b. water wrote:
 
> there was a recent thread on "responses to new users" that 
> may be relevant or of note to you to increase the likelihood 
> of having your queries answered.
> 
> i used stata 8.2 and had never before had to run zip or 
> zinb but a quick check on [R] did not show random effects 
> option, unless it was built-in into the command itself.
> 
> but "I was wondering if it was possible to run a zero 
> inflated negative binomial model on panel data with a random 
> effects specification." would come under "why not just run it 
> on your panel data and post the lists the outputs/problems 
> along with your queries" response. i am almost certain that 
> would elicit more response than the rather broad "Any 
> comments or suggestions..."
 
> >> From: sb4@u.washington.edu
> >> To: statalist@hsphsun2.harvard.edu
> >> Subject: st: Zero Inflated Negative Binomial Model on Panel Data?

> I posted this enquiry about a week back but did not get 
> any response. Hoping for better luck this time. I was 
> wondering if it was possible to run a zero inflated negative 
> binomial model on panel data with a random effects 
> specification. I tried the "zinb" command but did not get it 
> to consider group effects. 

*
*   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