Hi All,
Has anyone successfully estimated a multinomial probit model with
non-diagonal covariance matrix in WinBUGS? I would greatly appreciate it if
you could kindly share your code with me.
I searched the list archive and found some code provided by Andre Bonfrer,
which implements McCulloch & Rossi (1994) but restricts the covariance
matrix to be diagonal.
Since I don't think WinBUGS allows a Cholesky decomposition, I cannot
implement McCulloch & Rossi (1994). Instead, I tried to implement Albert &
Chib (1993), i.e. directly rejection-sampling the latent variables from the
multivariate normal distribution until the required constraints are
satisfied. This is supposed to be feasible in the new WinBUGS1.4. But it
takes FOREVER for the chain to move! Is slow updating due to inefficient
rejection sampling in this case?
Thank you very much for your attention.
Regards,
Yuanping
-------------------------------------------------------------------
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
|