Re: st: RE: 'nonest' option in 'xtreg'

[Nick Cox <njcoxstata@gmail.com>]

Good find. Where is it documented?

Nick

On 8 Jul 2011, at 13:26, Austin Nichols <austinnichols@gmail.com> wrote:

> Nick--
>
> From -help whatsnew9-:
> --update 15sep2005--
> (30)B.  xtreg, fe and xtreg, re produced cluster-robust VCEs when the
> panels were not nested within the clusters.  In some cases this VCE is
> consistent, and in others it is not.  You must now specify the new
> nonest option to get a cluster-robust VCE when the panels are not
> nested within the clusters.
>
> Without a detailed description of the data and the assumed correct dgp
> from the original poster, it will be hard for anyone to comment
> usefully  on the estimation problem.
>
> It strikes me as strange to model RE in one dimension and clustering
> of errors in another; two-dimensional clustering or a multilevel model
> seems a more natural starting point.
>
> On Fri, Jul 8, 2011 at 8:02 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> > I've just searched the .pdf of the entire [XT] manual and found no  
> > reference to "nonest".
> >
> > So, you know more than is in the manual.
> >
> > I think you got no answer partly because you are not telling us,  
> > exactly and precisely,
> >
> > 1. What you have typed into Stata.
> >
> > 2. What "limited" information means.
> >
> > 3. Where you found about this on Statalist.
> >
> > Nick
> > n.j.cox@durham.ac.uk
> >
> > Sejeong Ha
> >
> > I have not received any comment on my recent post. So, I am trying  
> > once more. I hope any of you can help. Thank you.
> >
> > My panel is an individual person and I am trying to cluster  
> > individuals by their household. A simple specification would look  
> > like
> >
> > xtset  individual year, yearly
> > xtreg  x y, cluster(household)
> >
> > However, I have an error message of "panels are not nested within  
> > clusters". With some examination through Statalist, I realised that  
> > my panels (individuals) are not in the same cluster (household) all  
> > the time (because individuals move in and out of households over  
> > time). Then, I've found that 'nonest' option somehow helps not  
> > receiving the same error. Though I wanted to find out more about  
> > this option, the information online seems limited. So, I'd like to  
> > ask you (1) what 'nonest' option does and (2) whether it is a right  
> > cure in
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/