Re: st: RE: cluster option and random effects

[Jeph Herrin <junk@spandrel.net>]

Is there a specific reason you want to use -cluster()- instead of
another approach? -xtmelogit- does not have a -cluster()- option,
but it will allow you to account for the clustering you describe
with the error structure. For instance

  xtmelogit depvar indvar || household: || person:

will model the main effect (intercept) as random across -person-s within
-household-s.

hth,
Jeph

Dimitrije Tišma wrote:
> I am sorry for not having been precise enough. What I would normally
> need is -xtlogit-, not xtreg. And, if I am not wrong, -xtlogit- won't
> allow me to cluster individuals on a higher level (households, in this
> case). Best, Dimitrije.
>
>
>
>
> 2010/1/2 Martin Weiss <martin.weiss1@gmx.de>:
> > <>
> >
> > If all you want is a random _intercept_, you should of course look at
> > -xtreg, re-, which does allow a -cluster()- option.
> >
> >
> > HTH
> > Martin
> >
> > -----Original Message-----
> > From: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Dimitrije Tišma
> > Sent: Samstag, 2. Januar 2010 16:29
> > To: statalist@hsphsun2.harvard.edu
> > Subject: st: cluster option and random effects
> >
> > Hi Statalisters,
> >
> > I have a very short yet possibly tough question. I need to estimate a
> > random intercept model using data with not entirely independent
> > observations within households. Basically, I would like to know what
> > possibilities there are in Stata to estimate random effects when using
> > cluster( ) option. I have already used -gllamm-, but the coefficients
> > of duration dependence variables do not the way I would expect them
> > to. So, I wanted to ask if there are there any alternatives to
> > -gllamm- (which I like and use a lot, but I really need to be
> > comfortable about my results and want to do double check)? Happy new
> > year to all, Dimitrije.
> > *
> > *   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/
> >
> >
> > *
> > *   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/
>
> *
> *   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/
*
*   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/