| T O P I C R E V I E W |
| helporigin8 |
Posted - 02/18/2022 : 09:58:21 AM
Origin Ver. and Service Release (Select Help-->About Origin): OriginPro 2018 64-bit
Operating System: Windows 10
Hi,
I am trying to fit my set of experimental data using Nonlinear Curve Fit option where I created Hankel function for momentum distribution to describe interaction between 5He and 7Li. However, when I get fit to the data, it seams that error bars that Origin calculates for the free fitting parameters are way to high regarding to the error bars on experimental points and regarding how experimental points are scattered.
Also, 95% confidence band seems to be way too wide.
Attached is the figure with mentioned fit and experimental data. |
| 3 L A T E S T R E P L I E S (Newest First) |
| easwar |
Posted - 02/21/2022 : 10:16:50 AM
Hello,
Sorry for the delay, we sent you an email from our ticket system, so please respond to that.
Easwar
OriginLab |
| helporigin8 |
Posted - 02/18/2022 : 5:47:34 PM
I got this message notifying me that fit succesfuly converged after 6 iterations:
(1) Parameter Initialization was called.
(2) ----------Levenberg-Marquardt-----------
Reduced Chi-sqr = 83.531630379
COD(R^2) = -1.2617205311643
Iterations Performed = 6
Total Iterations in Session = 6
(3) Fit converged. Chi-Sqr tolerance value of 1E-9 was reached.
Attached are my project file and the fitting function.
https://my.originlab.com/ftp/forum_and_kbase/Images/Hankel.FDFhttps://my.originlab.com/ftp/forum_and_kbase/Images/Sheet1a.dat |
| YimingChen |
Posted - 02/18/2022 : 11:49:36 AM
The R-square value is negative which indicates the fitting is not converged properly. Can you share your project file and the fitting function?
James |
