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]
st: interpretating xtmixed as decomposition
From
"Lim, Raymond" <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: interpretating xtmixed as decomposition
Date
Fri, 11 Nov 2011 16:57:18 -0500
Hello Statalister,
If I do xtmixed with two levels, can I interpret sd(_cons) as decomposition of unobserved
effects? For example, I run the command below. Student is nested in schools. Is it correct to
say of the unobserved random effects, 53% is due to school and 48% is due to student random
effects? I get 53% from .1298 / (.1298 + .1155)
xtmixed testscore controlvars | schoolid: | studentid:
The bottom of the output says
------------------------------------------------------------------------------
Random-effects Parameters | Estimate Std. Err. [95% Conf. Interval]
-----------------------------+------------------------------------------------
schoolid: Identity |
sd(_cons) | .1298335 .0071065 .116626 .1445366
-----------------------------+------------------------------------------------
studentid: Identity |
sd(_cons) | .1155506 .0031687 .109504 .1219311
-----------------------------+------------------------------------------------
sd(Residual) | .8789076 .0004965 .877935 .8798814
------------------------------------------------------------------------------
LR test vs. linear regression: chi2(2) = 37656.95 Prob > chi2 = 0.0000
---
Raymond Lim
Research Coordinator
Columbia Business School - Finance & Economics Division
[email protected]
*
* 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/