Thank you BUGS Fans,
I simply asked for an example of nonparametric density estimation using
WinBugs. I did not say so, but I was hoping to avoid binning the data
and modeling the counts by 'histogram smoothing,' yet also avoid a
'mixture of normal distributions'. Something similar to the
implementation of Kooperberg's logspline() in Splus would be perfect,
provided I could then incorporate it as part of a bigger model.
The answers from David Spiegelhalter and John Wilkinson were very
helpful, although they each leave me some points to work out.
David Spiegelhalter replied that program 6.27 on 'eye-tracking' on
http://www.mrc-bsu.cam.ac.uk/bugs/winbugs/examples.shtml
is a Dirichlet process prior model adapted from a Congdon example.
John Wilkinson sent a link to a paper by Lang, Adebayo, and Faahrmeir
on "Bayesian semiparametric seemingly unrelated regression"
http://www.quantlet.de/scripts/compstat2002_wh/paper/full/B_03_lang.pdf
and another by Smith and Kohn (1996, Nonparametric regression using
Bayesian variable selection. J. Econometrics 75:317-344). He also
refered me to section 4.4 (p.105 -109, "Nonparmetric Regression Via
Splines") in Congdon's 2001 book, specifically worked examples #4.8 and
#4.9 in the downloaded examples that accompany the book. Example 4.9 is
in fact an explicit model based on Smith and Kohn's work.
Thanks again to David and John.
Charles
>>> Charles Anderson <[log in to unmask]> 10/23/2003
1:21:24 PM >>>
Bugs Fans,
Would anyone have an example of nonparametric density estimation using
WinBugs that they could send me?
From the archives, there may be an example in Peter Congdon's book,
Bayesian Statistical Modelling (Wiley, 2001), especially pages 260 et
seq. I don't recognize an example in the online programs that
accompany
that book, and I've just bought his other book, so I'll first try for
some examples from this group.
Thank you for any help.
Charles
-------------------------------------------------------------------
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
-------------------------------------------------------------------
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
|