The reply below was inadvertenly omitted from the summary posted yesterday.
Cheers,
Arthur
------------------------------
Frederic Gosselin writes:
I have found out exactly the same problems as you for the use of Winbugs
on mixed effects models with moderate-size data (around 2,000 to 10,000
lines). As regards mixing, there are some techniques to make models mix
better. But as for speed, the different comments found in the mail
archive (cf.
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind03&L=BUGS&P=R28198&I=-3,
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0502&L=BUGS&P=R976&I=-3,
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind03&L=BUGS&P=R22345&I=-3;
but counter-examples are given in
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind0502&L=BUGS&P=R115&I=-3)
as well as others on the Internet (cf.
http://web.maths.unsw.edu.au/~wand/glmm2x2.pdf) make me think that for
these kinds of simulations winbugs is much too slow and whould be
replaced with programs in C/Fortran/C++ or Matlab (maybe also: R).
There are a varietry of "packages" that prepare the work to do so: among
them: MCMCpack in R (http://mcmcpack.wustl.edu/); and MC2Pack under Ox
(http://www.tinbergen.nl/~cbos/index.html?content=/~cbos/software/mc2pack.ht
ml) (but there are other ones). I am currently testing the speed of
these relative packs on the same case... If you are willing to
participate to this test, you are welcome...
----------------------------
*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*
Dr. Arthur M. Greene
International Research Institute for Climate Prediction (IRI)
The Earth Institute at Columbia University
202 Monell Building / 61 Route 9W / PO Box 1000
Palisades, NY 10964-8000 / USA
Tel: 845-680-4436 Fax: 845-680-4865
[log in to unmask]
*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*^*~*
-------------------------------------------------------------------
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
|