| T O P I C R E V I E W |
| sccv |
Posted - 12/18/2004 : 02:52:47 AM
Origin Version (Select Help-->About Origin):
Operating System: win XP
Origin version 6.1
Currently, I want to use Origin to fit the function as below:
y = y0 + A * exp(B * x / (C + x))
Where the value of C is known and C is a contant. I would like to find y0, A and B using fitting method in Origin. Could anybody help me? I am a beginner to Origin. |
| 4 L A T E S T R E P L I E S (Newest First) |
| easwar |
Posted - 12/21/2004 : 12:11:03 PM
quote:
By the way, do you have any idea to reduce the error after fitting?
Hi,
Not sure what you mean...the minimization process, if it converges, finds the best set of parameters for the function that represents your data. If the error on the parameters, at the end of this process, are large, that could be due to various things such as too much fluctuation in data, fitting function not appropriate for data, too many parameters in function etc.
Here is an article on curve fitting that could be of help in understanding the above:
http://www.tipmagazine.com/tip/INPHFA/vol-9/iss-2/p24.html
Easwar
OriginLab
|
| sccv |
Posted - 12/21/2004 : 03:41:52 AM
Thank you,
I got it. By the way, do you have any idea to reduce the error after fitting?
|
| easwar |
Posted - 12/20/2004 : 10:35:21 AM
Hi sccv,
You will need to define/create a new fitting function, which you can do with the advanced fitting tool. Look up help files on how to create a new fitting function.
As for using a constant, you can define constants in your fitting function by literally typing in the contant value in your expression, or also by defining a constant such as C in your case. To define constants to be used in fitting, see the Script->Parameter Initialization page of the NLSF tool.
Easwar
OriginLab
|
| lorentz |
Posted - 12/20/2004 : 08:12:43 AM
You should choose Analysis->Non-linear Curves fit->advanced fitting tool
then select the funciton you wanted and select the data you want to fit.
then choose fit from the fitting box.and it will ask you to initialize the parameter. |
|
|
