Statalist The Stata Listserver


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

st: RE: testing if distribution is negative binomial


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: testing if distribution is negative binomial
Date   Mon, 17 Jul 2006 17:06:00 +0100

None of this is necessary at all. 

You don't need to have covariates to use -nbreg-. 
Therefore, in principle the answer is just to 
use -nbreg-, as you are fitting a distribution, 
but without covariates.

In practice, there are various ways of parameterising
the negative binomial, and the parameterisation used
by -nbreg-, for good reasons, may differ from whatever
parameterisation you are accustomed to. 

This problem is again no problem, as -nlcom- makes it 
quite easy to map from one parameterisation to another. 

Alternatively, for example, -nbfit- from SSC by Roberto 
Gutierrez and myself is a worked example of code to use 
one parameterisation common in ecology, with a worked 
example in the help. 

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

Radu Ban
 
> I have a variable and I would like to test whether its distribution
> can be assumed to be negative binomial.
> 
> The one way I thought to do this, is to generate a true negative
> binomial distribution and to compare then the two distributions
> through a chi-squared test. However, I wasn't able to find a random
> number generator that gives a negative binomial distribution (the
> -rnd- that I found doesn't have this option). Does anyone have any
> suggestions?

*
*   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