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: Date: Thu, 7 Apr 2011 16:20:17 -0400
From
Daniel Marcelino <[email protected]>
To
[email protected]
Subject
Re: st: Date: Thu, 7 Apr 2011 16:20:17 -0400
Date
Thu, 7 Apr 2011 18:59:15 -0300
Luisa,
maybe I misunderstood part of your inquire. However, some words can
help you think more specific about your outcome.
First, yes you should interpret your coefficient as you do in a linear
regression. Regard that the main difference between both techniques is
the shrinkage and liability aspects, not the coefficient or slope
interpretation.
Second, as I understood from your question a more accurate model of
change might be the trajectory over time from identity variation.
Daniel
On Thu, Apr 7, 2011 at 5:20 PM, Luisa Soares-Miranda
<[email protected]> wrote:
> Dear All,
>
> I have one doubt and I think you can help me. I am a Phd student and I have conducted one mix-model analysis using the command xtmixed to analyze associations over 3 years of data. so I have xi: xtmixed .... || codigo: i.year, covariance(identity). So my question is can I interpret the coefficient as I interpret it in a linear regression? For example if I increase one unit in my exposure the outcome will change Coef. over time? Or I should only say that exposure and outcome are associated and in each direction, over time.
>
> thank you so much
> regards
>
> Luisa
> *
> * 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/
>
--
About me: http://danielmarcelino.zip.net/
*
* 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/