Could anybody provide algorithmic information on the calculation of the “MC
error” in WinBUGS? I have browsed the archive of this mailing list and
found several similar questions but no answer to them. IMHO such
information is really important since the tutorial of WinBUGS suggests
using the MC error as a rough indication of convergence (MC error < 5 % sd)
and a lot of academic courses and scientific papers make use of this value.
I think I have understood the semantics of this measure and I am aware of
the batch means method ([Roberts (1996; p.50)] pointed out by the on-line
help of WinBUGS) for calculating “sigma” in MC error = sigma/sqrt(N).
I have tried to write code according to these references and never obtained
the value of the MC error reported by WinBUGS (on samples generated by
WinBUGS). Since the choice of the batch size (the “m” parameter in [Roberts
(1996; p.50)]) is not specified in the above references, I have
calculated “sigma” for all “m” varying between 1 and N and never obtained
the value of “MC error” reported by WinBUGS.
-------------------------------------------------------------------
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
|