The following brief report was written for RSS News the news magazine for
members of RSS.
Practical Bayes using WinBUGS Workshop
On 22 October a nearly full lecture room attended the second Practical Bayes
using WinBUGS Workshop organised by the Business and Industrial Section of
the RSS. The 80+ attendees included delegates from France, Italy, Gran
Canaria and even the USA. It is also interesting to note that a fair
proportion were not professional statisticians. 17 of the 40 delegates who
returned the feedback form were not members of the RSS, with professions as
diverse as political scientist, fisheries scientist, physicist, health
economist, economist, engineer and corporate finance researcher listed.
Although it might have been the chance to sample the famous RSS refreshments
that attracted them, it was more likely the opportunity to meet David
Spiegelhalter, Nicky Best and Andrew Thomas - the creative force behind
WinBUGS - that made the £20/25 fee so attractive. One of the lessons from
the first workshop run in 1999 was the need to cater for a diverse audience
in terms of knowledge of the software. This was initially achieved by
running parallel sessions in which the novices (about 80% of the delegates)
listened to an introduction to WinBUGS by David Spiegelhalter, while the
more experienced users listened to Nicky Best and Andrew Thomas explain the
new features in soon to be released Version 1.4. It is clear from the
numbers attending the introductory session that there is a large and growing
interest in this software by scientists and statisticians working in a
variety of fields. For those who missed this day, or simply want to find
out more, should note that there will be introductory courses run prior to
RSS 2002 in Plymouth next September.
Much of the rest of the day was taken up by three case studies: Peter Blood
on "Using WinBUGS for some PK models", Stefan van Dongen on "Modeling the
unobservable developmental stability using a Bayesian latent variable model"
and Andrew Millard on "Logistic regression and missing data: estimating age
of human remains from toothwear." Copies of these presentations, as well as
the examples used by David in his introduction, can be downloaded from the
BUGS website http://www.mrc-bsu.cam.ac.uk/bugs : "Latest News". The
delegates were also given the opportunity to "ask the experts" in an open
forum session. Finally, we were given an insight into what the developers
were working on for future releases. It appears that as computing power
gets ever greater this software will continue to develop as a major tool for
the practical implementation of complex Bayesian models.
Once again the feedback, both verbally on the day and via the feedback
forms, provide valuable information for improving this type of meeting. It
was generally well received, but it is still evident that there are
difficulties to be overcome in targeting such a diverse audience. Anyone
wishing to find out more about the software should visit the website,
address given above, where it still can be downloaded free of charge.
PFIZER GLOBAL RESEARCH AND DEVELOPMENT
----------------------------------------------------------------
This message and any attachment has been virus checked by the
PGRD Sandwich Data Centre.
----------------------------------------------------------------
-------------------------------------------------------------------
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
|