Bookmark and Share

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: xtmixed


From   William Buchanan <william@williambuchanan.net>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: xtmixed
Date   Thu, 3 Jan 2013 16:35:40 -0800

Hi Carol,

You could not have two lines with identical slopes in your example.  In your specific example you do get two straight lines, but it isn't clear why this is a problem for you. In addition to running the syntax you provided, I also ran:

webuse pig
g drug=0 if id<=24
replace drug=1 if drug!=0
xtmixed weight c.week##i.drug || id: week, cov(uns) res(ind) mle variance
margins drug, at(week=(1(1)9))
marginsplot

You should also include the output from your original syntax since it is likely that the margins are fairly similar across the values of the independent variable.  Else, provide some additional details to the listserv regarding why you feel the line should not be straight along with an example that performs how you believe it should (in this case the pig weight dataset didn't produce two lines with identical slopes which you can see by looking at the difference in the values in the margins before the lines intersect and afterwards).

HTH,
Billy


On Jan 3, 2013, at 4:06 PM, Sun, Wensheng wrote:

> Hi, William,
> 
> Thanks for the reply. For example, 
> 
> webuse pig
> 
> gen drug=.
> replace drug=0 if id<=24
> replace drug=1 if id>24
> gen drug_week=drug*week
> xtmixed weight week i.drug drug_week, ||id: week, covariance(unstructured) residuals(independent) mle variance
> 
> I want the independent variable 'week' to be random in coefficient and intercept, eg. each id has its own slope and overall, different drug group has different slope at each week. However, if I use the above command, when I run
> 
> margins drug, at(week=(1 5 9))
> marginsplot
> 
> what I got is two straight lines with exactly the same slopes.
> 
> The marginsplot I wish to get is the ones in www.stata.com/stata12/margins-plots
> 
> Please help.
> 
> Thank you very much!
> 
> Carol
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index