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
 Error from Fitted Curves

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
Won88 Posted - 10/15/2024 : 11:54:59 PM
Origin Ver. and Service Release (Select Help-->About Origin): Origin 2015 (64-bit) Sr2 Academic


Greetings,

I would like to extrapolate some data with SD values from a fitted curve (image link below). Based on the fitted data, I'm looking to get the concentration (x-axis) at which there is 10% effect (y-axis). After fitting, in the FitNLCurve tab there are columns (X and Y) which are titled "Fitted Curves Plot". They seem to be the data associated with the curve. There are X and Y values associated with each point, but no error values. Is there a way to obtain error values based on the fitted data without having to graph each dataset individually and then take the SD manually?

Thank you!

https://my.originlab.com/ftp/forum_and_kbase/Images/Compound.doc
1   L A T E S T    R E P L I E S    (Newest First)
NicholasSupport Posted - 10/16/2024 : 7:00:10 PM
Hello Samantha,

Thank you for reaching out. This is an interesting question!

Since the fitted function defines one Y value for each X value, we are not able to directly calculate the error as we did with the input data (multiple X values for each Y value). Let's consider an alternative approach.

We will need to prepare two columns:

(A) A column containing the error of the input data:
In your original worksheet, select the data and use the "Statistics on Row(s)" button () to calculate the error (you can select your preferred measure of error in the "Quantities" tab).


(B) A column containing the fitted curve's prediction for (and ONLY for) the original data's X values
This can be generated by changing the parameters of the curve fitting device (click one of the green lock icons > "Change Parameters..."). Near the "Advanced" section we used earlier, go to "Find X/Y" and check "Find Y from X". Click Fit/Done and you will find a new sheet appears.
In column A, paste in the original data's X values.
Then, create Column C and fill it with the error column we made in part A. Click Column C's header to show the Mini Toolbar and designate this column as Y Error ().

Select all three columns to graph (or add to your previous graph). Now you can visually see how the data's level of variation looks along the fitted curve.



Let me know your thoughts or questions.

Best,

Nicholas G.
nicholas@originlab.com
OriginLab Technical Support Team

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