Author |
Topic |
|
Won88
1 Posts |
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 |
|
NicholasSupport
USA
57 Posts |
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 |
|
|
|
Topic |
|
|
|