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
 Reducing Chi-Square in Curve Fitting		  New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic

sina1366

7 Posts
Posted - 06/11/2014 :  11:15:27 AM  Show Profile  Edit Topic  Reply with Quote  View user's IP address  Delete Topic
Origin Ver.: Pro8
Hi
I'm fitting some curves with my own non-linear equation, but the chi-square is not reduced. In fact, my friend has reached to a chi-square of 64 (he was using origin 7), while I can not reduce it to less than 128 in the same fitting. What reasons could influence such difference? Why can't I reach to lower chi-square? Are there certain reasons?
Thank you very much.

ss

JacquelineHe

287 Posts
Posted - 06/12/2014 :  06:31:40 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Hi,

Compared with Origin7, Origin8 is with some improvement. But it should not lead the the problem as your description.
Do you set the same initial value to the parameters when you do fitting in two versions?
If the initial parameters are different, you could get different result.

So if you use the same function, the same initial parameters to fit the same dataset, but get the worse result in the newer version, please send your files to our Tech Support department (tech@originlab.com). Then we can try to reproduce on our machine and find out the problem.
1. Please send the files: Origin project (.opj), User-defined fitting function file(.fdf). (The fitting function is saved in the User Files folder)

2. Please send us your detailed version and serial number information. You can check it by selecting Help: About Origin in the menu.

Thanks
Jacqueline
OriginLab
Edited by - JacquelineHe on 06/12/2014 06:32:22 AM
Go to Top of Page

sina1366

7 Posts
Posted - 06/13/2014 :  04:09:27 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Dear Jacqueline
Thank you for your guidance,
But I don't know exactly about initial values that my friend chose. I think the initial values must be different. But how can I distinguish the proper initial values?
I agree that it really affects the fitting level, but is that possible to decrease chi-square from 128 to 64 by setting proper initial values?
Thanks

ss
Go to Top of Page

JacquelineHe

287 Posts
Posted - 06/19/2014 :  03:37:34 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Hi SS,

Cause we without your dataset and the fitting function, we are hard to help you to get the better initial values.

So I suggest you could send us your files. Do you mind to follow the last reply and send us your files?

Thanks
Jacqueline
OriginLab
Edited by - JacquelineHe on 06/19/2014 03:38:28 AM
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