Statalist


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

st: Multivariate Poisson - Correlation of Error Terms


From   "U.G. Narloch" <ugn20@cam.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   st: Multivariate Poisson - Correlation of Error Terms
Date   08 Jul 2008 19:44:37 +0100

Dear STATA-List users,

I have the following model:

y1 = X’ β1 + Z1’ γ1 + ε1
y2 = X’ β2 + Z2’ γ2 + ε2
y3 = X’ β3 + Z3’ γ3 + ε3
y4 = X’ β4 + Z4’ γ4 + ε4.

The dependent variables are count variables and X is a vector of explaining variables that is identical in each equation whereas Z includes equation-specific variables that differ from equation to equation.

I assume that the four count processes are related to one another, so that the disturbance terms should be correlated. To estimate these four equations in a multivariate model I follow the approach suggested at: http://www.stata.com/statalist/archive/2003-08/msg00226.html http://www.stata.com/statalist/archive/2004-09/msg00599.html.

First, I estimate each Poisson regressions separately and second I combine these results in a joint model via a Seemingly Unrelated Estimation (SUEST). Having done this, I would like to test if the error terms are really correlated, so that the count regressions cannot be estimated independent from each other.

For instance, the mvprobit command gives the correlation of the error terms between each equation and a likelihood ratio test for the correlation of error terms in the model. So, I am wondering if something like this could be done after the suest command.

As so far I was not able to figure out any method to do this, I would appreciate any contribution!

Thanks a lot!
Ulf Narloch





*
* 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