The Origin Forum
File Exchange
Try Origin for Free
The Origin Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ | Send File to Tech support
 All Forums
 Origin Forum
 Origin Forum
 Fitting and error parameters

Note: You must be registered in order to post a reply.
To register, click here. Registration is FREE!

Screensize:
UserName:
Password:
Anti-Spam Code:
Format Mode:
Format: BoldItalicizedUnderlineStrikethrough Align LeftCenteredAlign Right Horizontal Rule Insert HyperlinkUpload FileInsert Image Insert CodeInsert QuoteInsert List
   
Message:

* HTML is OFF
* Forum Code is ON
Smilies
Smile [:)] Big Smile [:D] Cool [8D] Blush [:I]
Tongue [:P] Evil [):] Wink [;)] Clown [:o)]
Black Eye [B)] Eight Ball [8] Frown [:(] Shy [8)]
Shocked [:0] Angry [:(!] Dead [xx(] Sleepy [|)]
Kisses [:X] Approve [^] Disapprove [V] Question [?]

 
Check here to subscribe to this topic.
   

T O P I C    R E V I E W
mej Posted - 11/02/2004 : 09:20:27 AM
Origin Version : 6.1 Pro
Operating System: wind 2000

Hi everybody,
I want to fit two processes with the following equations:
F1=ainf+(a1*exp(a*x)+a2*exp(-a*x))*exp(-b*x/2) with condition a1+a2=-ainf
and
F2=ainf*(1-K*(1/h1*exp(h1*x)-1/h2*exp(h2*x))) with condition 1/K=1/h1+1/h2

First, how to give to origin the second condition
and my big problem is that the error in parameters (value+/- error) h1 (4.5+/-1.2), K (~-14+/-6), a1 (-3+/-10), and a2 (-3+/-10)
are big despite the experimental curves are well fitted by these expressions


1   L A T E S T    R E P L I E S    (Newest First)
easwar Posted - 11/02/2004 : 09:29:47 AM
Hi,

Are you looking for how to simultaneously fit two datasets with two different equations that share parameters? You can do that in the NLSF tool by specifying two dependent variables, which in your case are F1 and F2, with independent variable being x, and the parameters being ainf, a1, a, a2, b, K, h1, and h2.

As for the second condition (1/K=1/h1+1/h2), that is a nonlinear constraint and currently Origin supports only linear constraints for parameters.

As for the error, even if the fit curve defines the data very well, the error on the parameters could be large due to other reasons. One such reason could be that you have too many parameters in your equation (over-parametrized equation), in which case the chi-sq surface is rather broad and so the minimum in the parameter space is rather broad leading to large parameter errors.

Easwar
OriginLab



Edited by - easwar on 11/02/2004 09:30:49 AM

The Origin Forum © 2020 Originlab Corporation Go To Top Of Page
Snitz Forums 2000