comp.lang.idl-pvwave archive: archive » Fitting with tabulated models

Home » Public Forums » archive » Fitting with tabulated models

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Fitting with tabulated models [message #69284]

Mon, 11 January 2010 13:42

ianpaul.freeley
Messages: 18
Registered: March 2007

Junior Member

I have a large set of models where I have two (but plan to expand)
degrees of freedom. The models are not directly compared to data, but
need to be multiplied by a low-order polynomial to take out
calibration issues and massaged in other ways. I have a function that
returns a value of chi^2 for a model given some data, but it is not
trivial so I would like to minimize the number of models I have to
compare.

My question is, what's the best method for finding the chi^2 minimum
given tabulated function values? Should I try to use mpfit and make
sure the minimum step size is large enough? Should I use Amoeba and
set the scale keyword? Or is there another standard way to search
tabulated models?

Report message to a moderator

[Message index]

Fitting with tabulated models

By: ianpaul.freeley on Mon, 11 January 2010 13:42

Previous Topic:	Re: How to avoid BSQ?
Next Topic:	pointer & structure

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Dec 01 03:25:09 PST 2025

Total time taken to generate the page: 2.48673 seconds