Hi list,
Thanks all who replied to my question. One of you kindly sent me a version of BRugs which runs with OpenBUGS 3.0.3. However, this actually didn't solve my problem. The same problem surfaced again. On closer examination, I found that the engine of BRugs and OpenBUGS standalone (shortened to OpenBUGS below) don't quite seem to be the same. I ran the same model with the same data, and used modelGenInits() to obtain initial values, which I checked were the same in both versions. However, the next update immediately gives different values. I believe in both versions, the only sampler used was 'slice', although I don't know how to check which sampler is used with BRugs. Basically, in BRugs, the updating stopped giving the error message:
update error for node <spec[3097]> algorithm slice error can not sample node too many iteraitions
after 3551 iterations.
whereas it ran smoothly in OpenBUGS.
Can anyone tell me how the sampling differs between OpenBUGS and BRugs?
and whether it is possible to set up BRugs so that it gives exactly the same results as OpenBUGS?
Any insight into why that error message occurred would also be very helpful.
Thanks,
Tim
-----Original Message-----
From: Mak, Timothy
Sent: 11 December 2009 13:29
To: [log in to unmask]
Subject: BRugs and OpenBUGS
Hi list,
As I understand it, BRugs is currently running on OpenBUGS 3.0.2, whereas the latest version of OpenBUGS available is 3.0.3.
I have something that doesn't run properly in BRugs but runs ok in OpenBUGS 3.0.3, so I'm hoping that BRugs will be able to run OpenBUGS 3.0.3 soon. Do you know when this will be?
Thanks,
Tim
-------------------------------------------------------------------
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
|