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
Username:
 Password:
Save Password
Forgot your Password? | Admin Options

 All Forums
 Origin Forum
 Origin Forum
 how to compare two fitting result		  New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic

elvis_xia

China
Posts
Posted - 01/02/2006 :  08:30:10 AM  Show Profile  Edit Topic  Reply with Quote  View user's IP address  Delete Topic
Origin Version (Select Help-->About Origin):
Operating System:
I use the same fitting function to fit a group of data by diffenent fitting method. One result is done by Origin and the other is not. Now I want to compare the goodness of the two results.
I think the parameter chi^2/DOF can finish the work. But I have two question:
1. is the fitting result the same one by Origin?
one way is: analysis- non-linear curve fitting
the other is: tools - polynomial fit
2.the definetion of chi^2/dof is ?
sum(data point-fuction point)^2/dof, where dof=data point-number of parameter?

Mike Buess

USA
3037 Posts
Posted - 01/02/2006 :  09:22:38 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
1. NLSF's Parabola fitting function with no additional constraints or weighting gives the same result as Tools > Polynomial Fit.

2. Correct for an unweighted fit. Otherwise multiply each term in the sum by the appropriate weight. Your expression for dof is correct.

Mike Buess
Origin WebRing Member
Go to Top of Page

elvis_xia

China
Posts
Posted - 01/02/2006 :  10:38:20 PM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Thank you for your reply.
But I have another two queations.
The first is, is it true that Origin can only fit data with error in one coordinate? I think it seems that during my investigation.
The other is how to give chi^2 with weights in both coordinates?
Furthermore, I found that the fitting curve derived from the following two is the same.(I use 7.5 edition)
(1) tools- polynomial fit- setting(choose error as weight and use reduced chi^2)
(2)analysis -NLSF- advanced fitting tool - use control to set weight.

Edited by - elvis_xia on 01/03/2006 07:22:23 AM

Edited by - elvis_xia on 01/03/2006 07:25:36 AM

Edited by - elvis_xia on 01/03/2006 07:26:34 AM
Go to Top of Page

minimax

348 Posts
Posted - 01/05/2006 :  06:50:52 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Hi Elvis,

Origin 7.5 does not support weights in both coordinates. And we will add weights in both coordinates into linear fit in Origin version 8.

quote:

Thank you for your reply.
But I have another two queations.
The first is, is it true that Origin can only fit data with error in one coordinate? I think it seems that during my investigation.
The other is how to give chi^2 with weights in both coordinates?
Furthermore, I found that the fitting curve derived from the following two is the same.(I use 7.5 edition)
(1) tools- polynomial fit- setting(choose error as weight and use reduced chi^2)
(2)analysis -NLSF- advanced fitting tool - use control to set weight.

Edited by - elvis_xia on 01/03/2006 07:22:23 AM

Edited by - elvis_xia on 01/03/2006 07:25:36 AM

Edited by - elvis_xia on 01/03/2006 07:26:34 AM


Max
OriginLab GZoffice
Go to Top of Page
  Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic
 New Topic  Reply to Topic
 Printer Friendly
Jump To:
The Origin Forum © 2020 Originlab Corporation Go To Top Of Page
Snitz Forums 2000