| Author |
 Topic  |
|
|
Pythe
Netherlands
Posts |
 Posted - 03/11/2004 : 10:04:44 AM
|
I am just starting with Origin (version 6.0) and I have some difficulties defining a new nonlinear fitting function.
What I want to do is the following:
I have a dataset consisting of y values versus x values, forming a sigmoidal-type graph. This system should be fitted with the following formulas:
dy/dx=P1*V1*(1-V1)*(x^2)
V1=V2/(V2+1)
V2=exp((P2/8.3145)((1/P3)-(1/x)))
However, I can't seem to get it right and obtain a sigmoidal plot. Can anybody help me and tell me how to define this function? |
|
|
ML
USA
63 Posts |
Posted - 03/12/2004 : 10:09:45 AM
|
Hi Pythe,
V1 depends on V2, and dy/dx depends on V1, so you certainly have to reverse the order of formulas, such that V2 is evaluated first, then V1, and finally dy/dx.
You also need to set the Form of the function to Y-Script.
ML
Edited by - ML on 03/12/2004 10:10:07 AM |
 |
|
|
Pythe
Netherlands
Posts |
Posted - 03/12/2004 : 3:05:41 PM
|
Hi ML,
Sorry, I messed the order of the equations up in this post. I think I have already put them in the correct order, but I am not sure as I am not on my work right now. I will check it first thing monday morning.
I don't have any experience with formula syntax in Y-script. I have tried to set the form as Y-script, but at the time it did not work. But I deduct from your answer that the syntax is OK for Y-script. My biggest concern was with the dy/dx on the left side of one of the equations, but I guess I did OK?
As I said, I will try first thing on monday and let you know if it works!
Thanks,
Pythe |
|
|
|
easwar
USA
1965 Posts |
|
|
Pythe
Netherlands
Posts |
Posted - 03/15/2004 : 05:35:06 AM
|
Hi,
I still couldn't get it right, but I settled with taking the derivative of the data first, and then fitting it with the equations, except using "y=" instead of "dy/dx=". This works fine, but I am a lazy person and wanted to fit right away on the raw data.
I will try some more to get it right, but if I don't succed, I'll just settle with the derivative data.
Thanks a lot,
Pythe |
|
|
| |
Topic |
|
|
|
help wanted on new fitting function!