| Author |
 Topic  |
|
|
javiers
Spain
6 Posts |
 Posted - 11/17/2008 : 10:57:06 AM
|
Hi colleagues,
I am trying to fit some data to an equation like:
g(y) = f(y)/(-x*R1)
I have a worksheet X Y. once is plotted It corresponds to f(y) I want to fit that f(y) function to a g(y) function and find R1 that gives the best fit.
any idea, comment will be welcomed.
thanks
|
|
|
liuxiaopi
China
Posts |
Posted - 11/19/2008 : 01:37:45 AM
|
Hi, Javiers,
I can not quite understand your question.
I try to summarize your fitting case as follows:
You have X Y data (your worksheet data):
x f(x)
.. ..
.. ..
You try to fit these data into the following model function:
f(x)/(-x*R1). (Note that the x variable.).
If f(x) is explicitly known, then you can just supply fitted function f(x)/(-x*R1).
However, if your case is as follows:
your data:
x y=f(x)
.. ..
.. ..
f(x) is not explicitly known.
And in fact you want to fit the measured data x and y into an equation g(x,y, R1)=const to obtain best fitted parameters R1. Then this is another issue, or in fact the so-called implicit function fitting.
Thanks,
Jack |
Edited by - liuxiaopi on 11/19/2008 01:41:30 AM |
 |
|
|
javiers
Spain
6 Posts |
Posted - 11/19/2008 : 03:28:37 AM
|
Hi liuxiaopi,
thanks a lot for reading and for your comments. As you said, I am in the second case the function is not explicitly known. So it would be great if you could give me some advice about implicit function fitting.
Thanks for your help
Javier |
|
|
|
liuxiaopi
China
Posts |
Posted - 11/21/2008 : 04:26:16 AM
|
Hi, Javiers,
Currently, Originlab 8.0 still can not support implicit function fitting. A further version of originlab is considering adding this feature.
There is a temp solution:
For example to fit an ellipse x^2 / a^2 + y^2 / b^2 = 1,
x, y is your measured data.
The temp solution is :
1) Define two independant variables: (x,y)
2) Define a fitting function x2 / a2 + y2 / b2 - 1
3) Supply dependant variable data: all zero
As we supply dependant data manually, I doubt if this way of fitting can produce meaning-ful or reliable result. Although, I do obtain reasonable fitted result on an ellipse with noise.
Please check more on google.... for implicit function fitting.
Hope this helps,
|
|
|
| |
Topic |
|
|
|
fitting data with: f(y)=f(y)/x*R1, best R1 value