[R] repeated values, nlme, correlation structures

Spencer Graves spencer.graves at pdf.com
Sun Nov 20 19:19:26 CET 2005


	  The pure Bayesian could do two things:

	  (1) Modify the nlme code to accept a prior specification, then add 
log(prior) to the log(marginal likelihood), so the function would then 
estimate the Baysian posterior mode, and then output a posterior 
specification in a format compatible with the input.  I have done 
similar things, but not exactly this, and I don't think it would be too 
hard.  The most difficult part would be just selecting a format to use 
as the primary specification of the prior / posterior.

	  (2) Markov Chain Monte Carlo, which is something I have yet to do.

	  Best Wishes,
	  Spencer Graves

Patrick Giraudoux wrote:

> Looks fine... and at least accessible to my current understanding and 
> capacity.  I wonder if this kind of problem/method would not make a pure 
> Bayesian very excited (I know one quite obsessional about it)... and 
> propose an alternate approach from there (though beyond my own skill)...
> 
> Thanks a lot again. Will do like that ASAP (means w.e.).
> 
> Patrick
> 

-- 
Spencer Graves, PhD
Senior Development Engineer
PDF Solutions, Inc.
333 West San Carlos Street Suite 700
San Jose, CA 95110, USA

spencer.graves at pdf.com
www.pdf.com <http://www.pdf.com>
Tel:  408-938-4420
Fax: 408-280-7915




More information about the R-help mailing list