help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

using nelder_mead_min


From: echemist
Subject: using nelder_mead_min
Date: Fri, 22 Apr 2011 07:36:22 -0700 (PDT)

Hi all,

I am new to Octave.  

What I wish to do is to fit some experimental data to a model using the
Nelder-Mead method.

I am using the root mean squared deviation (RMSD) as my merit function
(function to be minimized by Nelder-Mead).

I am trying to figure out how to use the function nelder_mead_min.  I tried
searching the forum, but haven't found the answer to my question.  I
apologize if I missed something in that search.

According to the information here:

http://octave.sourceforge.net/optim/function/nelder_mead_min.html

the parameters that nelder_mead_min requires are the function name (as a
string), a list of arguments, and an optional vector of control variables.

I am having trouble with the list of arguments.  What are these arguments? 
My general understanding of the Nelder-Mead method is that it is started
with N+1 values defining the initial vertices of the simplex in an
N-dimension space.  So, is this list of arguments that N+1 initial guesses
for that simplex?  If I need to provide the function with other parameters
(not to be minimized by Nelder-Mead), how can I do that?

Thanks for any help!

Jason




--
View this message in context: 
http://octave.1599824.n4.nabble.com/using-nelder-mead-min-tp3468080p3468080.html
Sent from the Octave - General mailing list archive at Nabble.com.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]