Skip to content

MCMC

7 messages · John Logsdon, Martyn Plummer, Douglas Bates +3 more

John Logsdon
# 
Hi

Does anyone know of any R coding/functions for MCMC approaches?  I am
currently using BUGS but I wonder if the bazaar has produced anything?  I
think I am pushing BUGS to it's limit and possibly past it at the moment.

John

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Martyn Plummer
#
On 16-Mar-00 John Logsdon wrote:
I don't think you will find anything as polished as BUGS, but the
components are available for you to write your own custom MCMC program.
You have to do a lot of work by hand that is normally done by BUGS, e.g.
specifying full conditionals, writing the functions for logical
nodes, recognizing conjugate distributions, tuning the Gibbs sampler, etc.

* libnmath contained within R provides plenty of functions for simulating
  random variables from commonly used distributions.  Brian Ripley has
  kindly modified this so that it can be built separately. Using this you
  can handle conjugate distributions.

* Wally Gilks has made publically available software for doing Adaptive
  Rejection Metropolis sampling which can be downloaded from the MRC-BSU
  web site.  Using this you can handle more difficult cases:
  http://www.mrc-bsu.cam.ac.uk/Software/arsregister.shtml
  Some tuning is required for optimal performance.

I have used this approach on a small problem, and have an increased
appreciation of BUGS for having done so.

Martyn
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Douglas Bates
# 
Martyn Plummer <plummer at iarc.fr> writes:
I don't know of any R code or packages presently available for MCMC.
However, it would be a very interesting project to design such code,
either for R or for omegahat.  The concept of a function closure in R
could be quite helpful.  I think Ross Ihaka and/or Robert Gentleman
wrote a paper once on how a Bayesian model would naturally be
expressed as a function closure.  

It took me a long time to understand what a function closure is and
why it would be useful (my thanks for Robert Gentleman and Duncan
Temple Lang for explaining it to me - repeatedly) but for applications
like this it is definitely worth the time to learn about it.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Jim Ianelli
# 
I have found that coding models using the C++ library
with automatic differentiation (AD Model Builder, www.otter-rsch.com)
allows for a useful implementation of MCMC that is adaptive and
has the added benefit of working from the posterior mode
and the initial estimate of the covariance matrix for developing
reasonable multivariate jumping rules.  This software
can also be made to interface with R (or Splus excel etc)
through use of dll's.

Jim Ianelli

-----Original Message-----
From: owner-r-help at stat.math.ethz.ch
[mailto:owner-r-help at stat.math.ethz.ch]On Behalf Of John Logsdon
Sent: Thursday, March 16, 2000 3:56 AM
To: r-help at stat.math.ethz.ch
Subject: [R] MCMC


Hi

Does anyone know of any R coding/functions for MCMC approaches?  I am
currently using BUGS but I wonder if the bazaar has produced anything?  I
think I am pushing BUGS to it's limit and possibly past it at the moment.

John

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._
-------------- next part --------------
A non-text attachment was scrubbed...
Name: James Ianelli.vcf
Type: text/x-vcard
Size: 426 bytes
Desc: not available
Url : https://stat.ethz.ch/pipermail/r-help/attachments/20000316/9e10e16b/JamesIanelli.vcf
A.J. Rossini
# 
DB> I don't know of any R code or packages presently available for
    DB> MCMC.  However, it would be a very interesting project to
    DB> design such code, either for R or for omegahat.  The concept
    DB> of a function closure in R could be quite helpful.  I think
    DB> Ross Ihaka and/or Robert Gentleman wrote a paper once on how a
    DB> Bayesian model would naturally be expressed as a function
    DB> closure.

Greg Warnes has some code to do this (R and Omegahat).  At some point,
it should be pretty general, I'm not sure what stage it is at right
now... 

best,
-tony
Gregory R. Warnes
#
On 16 Mar 2000, A.J. Rossini wrote:
AR>> >>>>> "DB" == Douglas Bates <bates at stat.wisc.edu> writes:
  AR>> 
  AR>>     DB> I don't know of any R code or packages presently available for
  AR>>     DB> MCMC.  However, it would be a very interesting project to
  AR>>     DB> design such code, either for R or for omegahat.  The concept
  AR>>     DB> of a function closure in R could be quite helpful.  I think
  AR>>     DB> Ross Ihaka and/or Robert Gentleman wrote a paper once on how a
  AR>>     DB> Bayesian model would naturally be expressed as a function
  AR>>     DB> closure.
  AR>> 
  AR>> Greg Warnes has some code to do this (R and Omegahat).  At some point,
  AR>> it should be pretty general, I'm not sure what stage it is at right
  AR>> now... 
  AR>> 
  AR>> best,
  AR>> -tony
  AR>> 
  AR>> -- 
  AR>> A.J. Rossini		       Research Assistant Professor of Biostatistics 
  AR>> Biostatistics/Univ. of Washington  (Th)	Box 357232   206-543-1044 (3286=fax)
  AR>> Center for AIDS Research/HMC/UW	  (M/F)	Box 359931   206-731-3647 (3693=fax)
  AR>> VTN/SCHARP/FHCRC		 (Tu/W)	Box 358080   206-667-7025 (4812=fax)
  AR>> rossini@(biostat.washington.edu|u.washington.edu|hivnet.fhcrc.org)
  AR>> http://www.biostat.washington.edu/~rossini
  AR>> 
  AR>> 

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Gregory R. Warnes
# 
Yes, in fact, I do have a nice MCMC tool for omegahat.  The only thing it
lacks to be fully functional is a java distribution library, and good
documentation.  I'll be writing up a tech report on it "in the near
future" (tm).

-Gregory
On 16 Mar 2000, A.J. Rossini wrote:
AR>> >>>>> "DB" == Douglas Bates <bates at stat.wisc.edu> writes:
  AR>> 
  AR>>     DB> I don't know of any R code or packages presently available for
  AR>>     DB> MCMC.  However, it would be a very interesting project to
  AR>>     DB> design such code, either for R or for omegahat.  The concept
  AR>>     DB> of a function closure in R could be quite helpful.  I think
  AR>>     DB> Ross Ihaka and/or Robert Gentleman wrote a paper once on how a
  AR>>     DB> Bayesian model would naturally be expressed as a function
  AR>>     DB> closure.
  AR>> 
  AR>> Greg Warnes has some code to do this (R and Omegahat).  At some point,
  AR>> it should be pretty general, I'm not sure what stage it is at right
  AR>> now... 
  AR>> 
  AR>> best,
  AR>> -tony
  AR>> 
  AR>> -- 
  AR>> A.J. Rossini		       Research Assistant Professor of Biostatistics 
  AR>> Biostatistics/Univ. of Washington  (Th)	Box 357232   206-543-1044 (3286=fax)
  AR>> Center for AIDS Research/HMC/UW	  (M/F)	Box 359931   206-731-3647 (3693=fax)
  AR>> VTN/SCHARP/FHCRC		 (Tu/W)	Box 358080   206-667-7025 (4812=fax)
  AR>> rossini@(biostat.washington.edu|u.washington.edu|hivnet.fhcrc.org)
  AR>> http://www.biostat.washington.edu/~rossini
  AR>> 
  AR>> 

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._