No, you need -nlogit- not -xtmelogit-: the word nesting means very
different things in these two models: In -nlogit- you can think of
a sequence of decisions where some decisions are nested in others,
e.g. you first decide between fancy or family restaurants, and
given that you chose a fancy restaurant you can now choose between Christophers and MadCows. This seems to represent your kind of
problem, except with an extra level of nesting. In -xtmelogit- the
nesting refers to the observations: e.g. students are nested in
classes, which are nested in schools, which are nested in districts,
etc. This is a multi-level problem, which does not seem to describe
your situation.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
--- On Tue, 28/7/09, Brendon Gerding <[email protected]> wrote:
> From: Brendon Gerding <[email protected]>
> Subject: Re: st: 3-Level nlogit
> To: [email protected]
> Date: Tuesday, 28 July, 2009, 1:09 PM
> Hi Jeph,
>
> Thanks for your reply, I read the documentation and it
> seems like
> that's what I'm needing. I'm fairly new to Stata though and
> this is
> the first project I am working on, would you mind giving a
> brief
> explanation as to how I would use the xtmelogit command?
> I've tried to
> figure it out but am not sure that what I'm doing is
> right...
>
> I need to regress employment status of individuals on
> various factors,
> and I suspect that there will be different factors working
> at
> different levels within the employment statuses. The
> breakdown of
> employment status is:
>
> Population: -> Not Economically Active
>
> -> Economically Active -> Unemployed
>
>
>
> -> Employed
> And then employed may be split further into Formally
> Employed and
> Informally Employed.
>
> Please could you briefly explain how I could set up this
> regression
> using the xtmelogit. The first draft of my honors
> dissertation is due
> in two weeks, and I am desperately needing these results!
>
> Thanks for your time,
> Brendon Gerding
>
> On Mon, Jul 27, 2009 at 1:41 PM, Jeph Herrin<[email protected]>
> wrote:
> > Look at -xtmelogit- documentation.
> >
> > hth,
> > Jeph
> >
> >
> > Brendon Gerding wrote:
> >>
> >> Hi,
> >>
> >> I am needing to perform a 3-level nested logistic
> regression. So far
> >> all the commands and examples I have seen only
> involve 2 levels.
> >> Please could anyone offer advice on how to set up
> the 3-level nlogit,
> >> or perhaps where I could look for examples of this
> kind of thing?
> >>
> >> Any help would be greatly appreciated.
> >>
> >> Many Thanks,
> >> Brendon Gerding
> >> *
> >> * 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/
