Re: st: model for fractional data with panel data

[Daniel Simon <dhs29@cornell.edu>]

ok. thanks a lot.
Daniel

At 01:41 PM 11/7/2007 -0500, you wrote:
> Probably.  You should run some simulations to see how well it does in
> a model like yours, though...
>
> Note that -poisson- with -robust- SEs also works well for such cases,
> if you are willing to assume that E(y)=ln(Xb).
>
> On 11/7/07, Daniel Simon <dhs29@cornell.edu> wrote:
> > Austin - just a quick follow-up question. In my case (I was not the
> > original poster on this thread), I do have a non-trivial number of cases
> > where y=0. Therefore, the glm approach with fixed effects seems appropriate
> > - yes? thanks. Daniel
> >
> *
> *   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/
*
*   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/