Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: mim: xtmelogit & mim: xtmixed


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: mim: xtmelogit & mim: xtmixed
Date   Wed, 22 Apr 2009 19:10:47 +0000 (GMT)

I have just a couple of comments:

the closing brace for your loop should be on a new line. 

The -mim- command probably won't work, look at the example I gave 
earlier to see how to make it work.

Getting model level characterics, especially p-values, is very 
tricky in Multiple Imputation. I would recommend against trying
to do that unless you really really really need it. If you realy
think you need than I would reccomend that you look at this 
presentation by Rose Medeirso at the last North American Stata
Users' Group meeting: http://ideas.repec.org/p/boc/fsug08/11.html

-- Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


--- On Wed, 22/4/09, Kris Anderson <andersok@reed.edu> wrote:

> From: Kris Anderson <andersok@reed.edu>
> Subject: Re: st: mim: xtmelogit & mim: xtmixed
> To: statalist@hsphsun2.harvard.edu
> Date: Wednesday, 22 April, 2009, 5:53 PM
> Sorry about that. I am new to Statalist and the protocol.
> 
> Here's what I tried to do:
> 
> 
> mim: xtmixed avgdrink age sex gradedich white hisp affil
> socanx saaff_int if currentdrinker > 0, || mid_sch:,
> covariance(independent)
> 
> forvalues i = 1/35 {
> xtmixed avgdrink age sex gradedich white hisp affil socanx
> saaff_int || mid_sch:, covariance(independent) if
> _mj==`i'}
> 
> I can get estimates for the integrated model but I am
> trying to see the models for all 35 iterations in order to
> estimate overall Rs and p-values for the models. I have done
> this successfully with standard regressions but it stalls
> after the mim line here.
> 
> Maarten buis wrote:
> > --- On Wed, 22/4/09, Kris Anderson wrote:
> >> I am trying to run mim: xtmelogit and am running
> into
> >> errors. Has anyone successfully run this
> application?
> > 
> > *------------------ begin example
> -------------------------
> > webuse bangladesh, clear
> > xtmelogit c_use urban age child* || district:
> > 
> > replace urban = . if runiform() < invlogit(-2 +
> c_use)
> > replace age = . if runiform() < invlogit(-2 +
> c_use)
> > 
> > ice c_use urban age child*, m(5) clear
> > 
> > mim, cat(fit) : xtmelogit c_use urban age child* ||
> district:
> > *------------------ end example
> ---------------------------
> > 
> > Notice that I am not claiming that this imputation
> model
> > (the -ice- command) which ignores the nested structure
> is appropriate --- it probably isn't --- it is just
> there
> > to illustrate the mechanics of how to specify the
> -mim-
> > command in combination with -xtmelogit-.
> > 
> >> Also, I have successfully run mim: xtmixed but
> cannot get
> >> the independent model estimates to run. I saw a
> post
> >> sometime ago with a similar problem but no
> solution posted.
> >> Has anyone solved this issue?
> > 
> > I don't understand this question, what do you mean
> with
> > independent model (it could mean many things), and
> what
> > earlier post are you referring to? 
> > More generally, you could have made it a lot easier
> for us to answer your question by stating what you typed,
> what you expected as output, and what the actual output was.
> Statements like "running into errors" don't
> help much in trying to nail down what the problem is. 
> > Hope this helps,
> > Maarten
> > 
> > -----------------------------------------
> > Maarten L. Buis
> > Institut fuer Soziologie
> > Universitaet Tuebingen
> > Wilhelmstrasse 36
> > 72074 Tuebingen
> > Germany
> > 
> > http://home.fsw.vu.nl/m.buis/
> > -----------------------------------------
> > 
> > 
> > 
> > 
> >       
> > *
> > *   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index