Thankyou to the couple of hundred people who registered and downloaded
ELICITOR! (original email below).
Unfortunately there was a small typo in the WBDev piecewise function
which (in most cases) would still run but give wrong answers. Version
3.2 corrects this with a few other small changes, and also has the
source code for the 'link' and 'prior' modules so that alternative link
functions and prior distributions can be programmed in (please email me
for further info if you want more help on this). This is only available
in the add-on version and requires BlackBox (in the same way WBDev
does). Elicitor V3.2 is available without further registration(!) from
www.maths.qut.edu.au/~whateley
I am interested in gauging the interest in a 'elicitation' internet
group - for those using and developing elicitation techniques and those
interested in developing additional elicitation software components. I
am also interested in developing a 'wishlist' for elicitation
software...I have created an msn group at http://groups.msn.com/Elicitor
so that the winBUGS list is not abused for this purpose. So please
indicate you interest there - if there is no interest then the group
should delete itself in a couple of months!
Lastly, I would be very interested in feedback if anyone has actually
used the software.
Thanks,
Mary Kynn
____________________________________________________________________
Dear BUGSusers,
As a part of my PhD on eliciting expert prior distributions, I have
developed interactive, graphical elicitation software for the logistic
regression model. The software comes in two versions: a standalone
program (similar look and feel to winBUGS), or as a winBUGS add-on
component. Simply extract the zipped file to the winBUGS directory and
new menu items will appear next time winBUGS is started. There are also
additional modules for visualising probabilities in a variety of formats
and graphically displaying the parameters of a logistic regression
equation as probabilities on bar-charts and species response curves,
with credible intervals. There is a complete user manual with
walkthrough example, sample worksheet and interview questions.
www.maths.qut.edu.au/~whateley
_____________________________________________________
ELICITOR is a computer program that has been developed for eliciting
expert opinion as prior distributions for a Bayesian model, in a
standard menu-oriented Windows environment. It is designed to be user
friendly to the non-statistician, and to minimise common biases in
assessing probabilities where possible.
The scheme is graphical, interactive and requires no formal knowledge of
regression modelling or probability theory. Complete specification of
the model is not necessary before eliciting the prior distributions,
which allows the model to be built gradually.
ELICITOR produces the written model solely from the graphs and outputs
the model in standard notation, as an S-plus function for making
predictions and winBUGS code for a posterior analysis. (Bayesian
inference Using Gibbs Sampling is computer software for the Bayesian
analysis of statistical models using Markov chain Monte Carlo (MCMC)
methods (Spiegelhalter et al, 1999).)
Currently the program supports normal priors for a generalized linear
model with logit link. Also included are additional modules for
visualising probabilities and 'reverse engineering' a logistic
regression equation.
_______________________________________________________
-------------------------------------------------------------------
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
|