Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: How to plot the mean trajectory and 95%CI band afte mi estimate: xtmixed


From   Khady Brumblay <kndaobr@me.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: How to plot the mean trajectory and 95%CI band afte mi estimate: xtmixed
Date   Thu, 19 Jul 2012 13:45:06 -0400

Hi all,
I am trying to predict the average trajectories of a health outcome (health) overtime (year since an event). My predictors include the variables year (centered), year squared, as well as demographic (race, age, sex) and health characteristics (self-rated health, comorbid conditions). The data was imputed for item missing, using the stata mi command, which resulted in 5 datasets in addition to the original dataset. The data is currently in wide format, meaning each observation for each given year has one entry.

I have been unsuccessfully trying the plot the mean trajectory and 95%CI band after using [mi estimate: xtmixed]. I have tried different things, but none seem to work.
1) the twoway approach proposed by Rabe-Hesketh and Skrondal (p220) does not apply to imputed data using mi, and does not include time squared, interactions at different levels of data, or time varying-covariates.
2) the margin option suggested in the statalist does not work with [mi estimate] 
3) HLM does this fairly simply, and I am not sure why stata does not
*
*   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   |   Site index