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
 Curve fitting data with x error bounds
 New Topic  Reply to Topic
 Printer Friendly
Author Previous Topic Topic Next Topic Lock Topic Edit Topic Delete Topic New Topic Reply to Topic

ramcd

USA
4 Posts

Posted - 12/14/2014 :  7:19:39 PM  Show Profile  Edit Topic  Reply with Quote  View user's IP address  Delete Topic
I have location vs time data and want to obtain speed by curve fitting. The measured time has uncertainty while the locations do not. The data with time on x-axis and locations on y-axis seem to follow an exponential relation. So I am trying to curve fit an exponential curve keeping the time data on x-axis and locations on y-axis. How do I include the error in time (which is on x-axis) in the fitting routine?


rcd

Shirley_GZ

China
Posts

Posted - 12/14/2014 :  10:24:20 PM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Hi,

Since Origin 9.1, OriginPro supports ODR (Orthogonal Distance Regression), you can use it to do fitting with X error.



Any problem, please let me know.

Thanks,
Shirley

Originlab Technical Service Team
Go to Top of Page

ramcd

USA
4 Posts

Posted - 12/14/2014 :  11:17:27 PM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Hello,

Thank you for helping out. I tried using the ODR algorithm to fit an exponential curve- firstly by not including the error bounds, and secondly by including the error bounds. However, I am getting exactly same curve fits, statistics of fit, etc. While neglecting the errors, I select 'no weights' option in the Data Selection menu of the NLfit program. On the other hand, to include the errors I first set the error column as 'x-error', then select the 'Use Each Range Setting' option in NLfit. I run the curve fitting routine, but get exactly same fits and statistics in both the cases. Am I missing out on something?

To help you, I attache a snapshot of my data in OriginPro

rcd
Go to Top of Page

ramcd

USA
4 Posts

Posted - 12/14/2014 :  11:34:54 PM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Just an update, the result doesn't change even if I vary the errors in column C. It seems that the error bounds are somehow not getting included in the fitting routine...

rcd
Go to Top of Page

Shirley_GZ

China
Posts

Posted - 12/15/2014 :  12:42:36 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
Please tell me which function you are using to fit the data. Or send us the project via tech@originlab.com and quote this post.

Originlab Technical Service Team
Go to Top of Page

ramcd

USA
4 Posts

Posted - 12/15/2014 :  01:12:45 AM  Show Profile  Edit Reply  Reply with Quote  View user's IP address  Delete Reply
quote:
Originally posted by Shirley_GZ

Please tell me which function you are using to fit the data. Or send us the project via tech@originlab.com and quote this post.

Originlab Technical Service Team



sent the project file

rcd
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