Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: margins and xtmixed
From
Ricardo Ovaldia <[email protected]>
To
[email protected]
Subject
Re: st: margins and xtmixed
Date
Wed, 14 Mar 2012 10:35:53 -0700 (PDT)
David Hoaglin asked:
>I am not sure that I understand the random-effects part of your model.
>For each id it specifies a random intercept. Also, for each id it
>specifies a random coefficient for group. Please say more about the
>structure of your data. How is id related to group (and to time)?
Patients were randomly allocated to three treatment groups. Each patient was followed for a year and measurements made at baseline and every three months. Interest is on the effect of treatment. The model allows for a random intercept and I also allow for each treatment to have a different slope (i.e. the -||Id:group- part of the syntax statement). I am also allowing for and interaction between time (month and group).
Ricardo
--- On Wed, 3/14/12, Ricardo Ovaldia <[email protected]> wrote:
> From: Ricardo Ovaldia <[email protected]>
> Subject: st: margins and xtmixed
> To: "Statalist" <[email protected]>
> Date: Wednesday, March 14, 2012, 8:24 AM
>
> I am using –margins- and –test- after –xtmixed- to
> test if “groups” differ at specific times. Time is
> discrete in this study.
> I am concerned with the small SE that margin reports
> (delta-method adjusted). Small differences are found
> significant.Any help would be appreciated
>
> Here is an example:
>
> I fit the following -xtmixed- model:
> . xtmixed ft i.group##month || id: group, cov(unstructured)
> mle
>
> And I am interested in comparing the group effect at certain
> time, 3 months for example, I used –margins, post- and
> then –test-:
>
> . margins, over(group month) post
>
> Predictive margins
>
> Number of obs =
> 344
>
> Expression : Linear prediction, fixed
> portion, predict()
> over : group month
>
>
> ------------------------------------------------------------------------------
> |
> Delta-method
> |
> Margin Std. Err.
> z P>|z|
> [95% Conf. Interval]
> -------------+----------------------------------------------------------------
> group#month |
> 1 3
> | 6.382194 .0760204
> 83.95 0.000
> 6.233196 6.531191
> 1 6
> | 7.693377 .0819562
> 93.87 0.000
> 7.532746 7.854008
> 1 9
> | 8.709543 .1006514
> 86.53 0.000
> 8.51227 8.906816
> 1 12
> | 9.167792 .1323406
> 69.27 0.000
> 8.908409 9.427175
> 2 3
> | 6.053284 .0700887
> 86.37 0.000
> 5.915913 6.190655
> 2 6
> | 7.900573 .0777758 101.58 0.000
> 7.748135 8.053011
> 2 9
> | 9.124052 .1023024
> 89.19 0.000
> 8.923543 9.324561
> 2 12
> | 10.04959 .1374368
> 73.12 0.000
> 9.780223 10.31897
> 3 3
> | 6.074073 .0780239
> 77.85 0.000
> 5.921149 6.226997
> 3 6
> | 7.949809 .0856802
> 92.78 0.000
> 7.781879 8.117739
> 3 9
> | 9.337498 .1128377
> 82.75 0.000
> 9.11634 9.558655
> 3 12
> | 10.14703 .1480792
> 68.52 0.000
> 9.856798 10.43726
> ------------------------------------------------------------------------------
>
> . test 1.group#3.month= 2.group#3.month= 3.group#3.month
>
> ( 1) 1bn.group#3bn.month - 2.group#3bn.month = 0
> ( 2) 1bn.group#3bn.month - 3.group#3bn.month = 0
>
> chi2( 2)
> = 12.22
> Prob > chi2
> = 0.0022
>
> Is this correct?
>
> Thank you,
> Ricardo
>
> Ricardo Ovaldia, MS
> Statistician
> Oklahoma City, OK
>
>
>
> *
> * 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/