AnsweredAssumed Answered

AW: vrf More on nonlinear curve fitting w/Matlab script in Vee

Question asked by VRFuser on Aug 2, 2007
Thanks,
High order polynomials can provide a good fit but do not allow comparison with a model or extraction of values from that model- but they do provide an empirical equation that does describe the data-  they also tend to oscillate a bit much in some cases. 

For the EQ I am showing in my example a polynomial would fit well but no so well for exponential decay.


I did forget to mention that I set VEE to 18 digits of precision- probably don't need that much  but if too few the matrix inversion will probably fail.

When I get some time I plan to make the program more user friendly by doing numerical derivatives and an easier entry of the model equation, which has to be placed in twice in this version.
Jim



Baranski, Detlef wrote:
>  Hi Jim, very good job !
>
> But one question: Why was the regression object not good enough- especially when using Polynoms of higher order?
>
> -----Ursprüngliche Nachricht-----
> Von: Jim McClymer [mailto:McClymer@maine.edu]
> Gesendet: Montag, 30. Juli 2007 16:53
> An: VRF
> Cc: VRF
> Betreff: Re: [vrf] More on nonlinear curve fitting w/Matlab script in
> Vee
>
> Hi,
> I was unable to find any decent nonlinear curve fitting code for VEE and the included MATLAB function FMINSEARCH did an extremely poor job so I wrote my own Levenberg-Marquardt algorith in VEE. It seems to work very well in my testing- this version does require analytic derivatives and requires recoding the function.
> I hope you find it useful.
> Jim
>
> --
> Jim McClymer, PhD
> Graduate Coordinator
> Department of Physics and Astronomy
> University of Maine
>
> Der Inhalt dieser E-Mail (inkl. aller Anlagen) ist vertraulich und ausschließlich für den Adressaten bestimmt. Jede Art der Verbreitung, Nutzung oder Vervielfältigung ist untersagt. Sollten Sie diese E-Mail irrtümlich erhalten haben, informieren Sie bitte sofort den Absender und löschen die E-Mail.
>
> Unternehmensangaben der Gesellschaften mit Sitz in Deutschland:
>
> Pilkington Holding GmbH, Haydnstraße 19, 45884 Gelsenkirchen
> Sitz: Gelsenkirchen, Amtsgericht Gelsenkirchen HRB 8997
> Geschäftsführung: Jochen Settelmayer (Vorsitz), Dr. Clemens Miller,
> Thomas Kretschmann, Robert Hales Vorsitzender des Aufsichtsrates: Dr.
> Axel Wiesener
>
> Pilkington Deutschland AG, Haydnstraße 19, 45884 Gelsenkirchen         
> Sitz: Gelsenkirchen, Amtsgericht Gelsenkirchen HRB 2707
> Vorstand: Dr. Clemens Miller, Jochen Settelmayer Vorsitzender des
> Aufsichtsrates: Dr. Axel Wiesener
>
> Pilkington Automotive Deutschland GmbH, Otto-Seeling-Straße 7, 58455
> Witten
> Sitz: Witten, Amtsgericht Bochum HRB 8443
> Geschäftsführung: Thomas Kretschmann
> Vorsitzender des Aufsichtsrates: Jochen Settelmayer

> Bauglasindustrie GmbH, Hüttenstraße 33, 66839 Schmelz / Saar
> Sitz: Schmelz / Saar, Amtsgericht Saarbrücken HRB 52020
> Geschäftsführung: Manfred Ebbers
>  

--
Jim McClymer, PhD
Graduate Coordinator
Department of Physics and Astronomy
University of Maine


---
You are currently subscribed to vrf as: hua_jing@agilent.com
To subscribe please send an email to: "vrf-request@lists.it.agilent.com" with the word subscribe in the message body.
To unsubscribe send a blank email to "leave-vrf@it.lists.it.agilent.com".
To send messages to this mailing list,  email "vrf@agilent.com". 
If you need help with the mailing list send a message to
"owner-vrf@it.lists.it.agilent.com".
Search the "unofficial vrf archive" at "http://www.vrfarchive.com/vrf_archive".
Search the Agilent vrf archive at "http://vee.engineering.agilent.com".

Outcomes