>I couldn't help replying to this message with this suggestion since I think
>it's a crime that the Nelder-Mead routine is not in S-plus.
There is a version of the Nelder-Mead algorithm (essentially S from C from Press
et al "Numerical Recipes in C") included with my time series library "DSE"
available at <http://www.bank-banque-canada.ca/pgilbert>. The code is
independent from the time series parts of the library but has not been package
separately. It does suffer from S "for loop" problems, but I've incorporated a
way to work around that by saving after a number of iterations. I have not used
the code much in a few years, but it was once working fairly well.
Paul Gilbert
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|