Hello all,
I am working on fitting some models where there is a multivariate normal
distribution of level 2 effects. I'd like to constrain the covariance
matrix is what's essentially an arbitrary fashion that corresponds to
certain identification constraints. For instance, I often need to create
a matrix with the structure like this (for four random effects):
(1 )
(0 s22 )
(0 s32 s33 )
(0 s42 s43 1 )
So it's a 4X4 covariance matrix which would ordinary have ten free
parameters, but it only has five. The basic structure is the same for
more random effects. Another one would be
(1 )
(r21 1 )
(r31 r32 1 )
(r41 r42 r43 1 )
with the constraint that r41 + r42 + r43 = 1. (Again, this structure can
be generalized but the essential idea is expressed by the example.)
Anyone have a good reference or example?
Thanks,
Jay
--
J. Verkuilen
Department of Psychology
University of Illinois Urbana-Champaign
603 E. Daniel, MC-716
Champaign, IL 61820
[log in to unmask]
"If you've been playing poker for half an hour and you still don't know who the patsy is, you're the patsy." --Warren Buffett
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|