Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: supply known-variance matrix for xtmixed?

From   Brian Goodness <[email protected]>
To   [email protected]
Subject   st: supply known-variance matrix for xtmixed?
Date   Tue, 30 Aug 2011 17:12:29 -0700

Is there a way in STATA to supply a known-variance matrix (rather than
it be estimated) to the xtmixed command (i.e., specify G, the
covariance matrix for the random effects) for meta-analyses?  I am not
familiar with SAS, but according to its syntax for the mixed effects
specification (PROC MIXED), you can read in the G matrix by using the
GDATA= option in the RANDOM statement (see

Just wondering if this is possible using STATA, since I'm a STATA
user.  Any advice is much appreciated.


*   For searches and help try:

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