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

Re: st: seemingly unrelated negative binomial

From   Maarten buis <>
Subject   Re: st: seemingly unrelated negative binomial
Date   Tue, 31 Jul 2007 21:51:56 +0100 (BST)

--- Rachel Whaley <> wrote:
> I have four dependent variables (different count variables drawn from
> the same sample).  Ideally I would like to use SUR rather than four 
> separate regression analyses to control for the correlated errors and
> to compare the equality of coefficients across equations (f-tests). 
> However, the data are counts and require negative binomial
> regression. At this point, I have estimated the equations separately 
> using nbreg and I will calculate by hand a statistic to examine the
> equality of coefficients across models (the models are exactly the 
> same, just the dependent variables are different).  Unless there is a
> way to do this?

Below I give a substantively non-sensical example which explains the
mechanics of using -suest-: You estimate the equations seperately,
store each equation directly after you estimate it with -est store-,
and than use -suest- on the stored equations. I have used -poisson- as
-nbreg- did not converge quickly in this case (which is not surprising
since it is non-sensical), but in your case you can just as easily
replace -poisson- with -nbreg-.

*------------ begin example ------------
sysuse nlsw88, clear
poisson age married
est store A

poisson grade married
est store B

suest A B
test [A_age]married=[B_grade]married
*------------- end example --------------
(For more on how to use examples I sent to the Statalist, see )

Hope this helps,

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for
your free account today* 
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index