Re: st: invoking -test- for nested nbreg models with robust standard errors

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 04:37 PM 4/15/2004 -0700, Kjersten Bunker Whittington wrote:
> Hi Statalisters,
> I am running nested nbreg models, clustered on the id of the 
> respondent.  I would like to test whether my full model improves 
> satisfactorily upon my constrained model.  I realize that I cannot use 
> -lrtest- to do this, and I have read the Stata FAQ on using the -test- 
> command.  My question is one of logistics - what is the syntax I use to 
> achieve this task?  The way I would do it with lrtest is the
> following:
>
> nbreg y x1 x2 x3 x4, cluster(id)
> est store A
> nbreg y x1 x2 x3, cluster(id)
> lrtest A
>
> which we know is incorrect and doesn't work.   Can someone tell me the 
> equivilant commands using the -test- command?  I can't figure out how to 
> take into account multiple models with the test command, as it appears to 
> work off of only one model.
I have a feeling that I am missing something here and this is a stupid 
answer -- but wouldn't it just be

nbreg y x1 x2 x3 x4, cluster(id)
test x4

If, by some chance, the above is correct, I believe the chi-square 
statistic will be the square of the z statistic for x4.



-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW (personal):    http://www.nd.edu/~rwilliam
WWW (department):    http://www.nd.edu/~soc

*
*   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/