comp.lang.idl-pvwave archive: archive » Re: Gamma Variate Curve Fitting

Home » Public Forums » archive » Re: Gamma Variate Curve Fitting

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

Re: Gamma Variate Curve Fitting [message #35515 is a reply to message #35514]

Mon, 16 June 2003 10:07

Paul Van Delst[1]
Messages: 1157
Registered: April 2002

Senior Member

Kate wrote:
>
> I have been trying for a week now to fit data to a gamma variate curve
> using the curve fitting algorithms built into IDL. Every time I use
> CurveFit it tells me my data does not converge and every time I use
> SVDFIT my program crashes and exits out of IDL automatically! I made
> sure that my inital guesses were VERY close to the actual test data I
> am using and I'm still stuck. Does anyone have any experience fitting
> this type of non-linear curve using IDL? If I was using anything else
> I would probably use the simplex method but there doesnt' seem to be a
> downhill simplex curve fitting algorithm in IDL. HELP!!!

Go to Craig Markwardt's website and download his MPFIT fitting library:
http://cow.physics.wisc.edu/~craigm/idl/fitting.html

I had a similar problem recently (and I was just fitting a relatively simple polynomial
[non-integer exponents] but I tried all the usual suspects) and when I simply replaced
IDL's CURVEFIT with Craig's MPCURVEFIT all my problems went away...literally. Where
CURVEFIT didn't converge, the MPFIT equivalent did. And where CURVEFIT did converge, the
MPFIT equivalent gave me a better fit! And when I really screwed up the initial fit
coefficient guesses and MPCURVEFIT didn't converge, I got meaningful error output that
allowed me to fix the problem.

If I sound gushy about the MPFIT stuff it's because I am - using that code saved me monthS
of work, and I'm not exaggerating.

I hereby nominate Craig and his software for a special award for services to the
school...I mean, the IDL community. :o)

paulv

--
Paul van Delst
CIMSS @ NOAA/NCEP/EMC
Ph: (301)763-8000 x7748
Fax:(301)763-8545

Report message to a moderator

[Message index]

		Re: Gamma Variate Curve Fitting By: faeriepunk on Thu, 19 June 2003 06:55
		Re: Gamma Variate Curve Fitting By: Craig Markwardt on Wed, 18 June 2003 08:34
		Re: Gamma Variate Curve Fitting By: Andrew Meigs on Wed, 18 June 2003 02:20
		Re: Gamma Variate Curve Fitting By: robert.dimeo on Mon, 16 June 2003 14:41
		Re: Gamma Variate Curve Fitting By: Paul Van Delst[1] on Mon, 16 June 2003 10:07

Previous Topic:	Re: sec : U Re: Coordinates from SLIDE_IMAGE
Next Topic:	Function Maximum

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sun Oct 12 04:45:42 PDT 2025

Total time taken to generate the page: 8.47986 seconds