The experimental data (y value) in esfit are 1D array, that means lost of x coordinate of the data.
If the space between x coordinates were different (e.g., 2 4 6 8 10 12 14 16 18 20 30 40 50 60 70...),
rescaling will contort the experimental plot (the x coordinates seems to be 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15...).
Of course the fitting doesn't be correct.
How can I overcome this problem?
Thanks a lot!
A rescaling problem of esfit
-
Matt Krzyaniak
- EasySpin Guru
- Posts: 153
- Joined: Tue Jul 22, 2014 11:01 am
- Location: Northwestern University
Re: A rescaling problem of esfit
Rescaling of the data through
At the moment esfit doesn't take into account the x coordinate of the experimental data for a couple reasons: your simulation should have its x coordinate defined such that it will reproduce the experimental x coordinate, for least square fitting the x coordinate is irrelevant all that is important is the difference between the experimental data and the simulation.
In general the ES simulation routines do not support nonlinear abscissa, so if you're trying to use a nonlinear one you'll need to write a custom function which takes that into account.
'lsq' or 'lsq0' should be independent of the x coordinate and therefor not contort your experimental plot at all. At the moment esfit doesn't take into account the x coordinate of the experimental data for a couple reasons: your simulation should have its x coordinate defined such that it will reproduce the experimental x coordinate, for least square fitting the x coordinate is irrelevant all that is important is the difference between the experimental data and the simulation.
In general the ES simulation routines do not support nonlinear abscissa, so if you're trying to use a nonlinear one you'll need to write a custom function which takes that into account.
Re: A rescaling problem of esfit
I don't understand whether your rescaling problem was on the x or y axis. I have had a similar issue when fitting magnetic susceptibility data and I had a distortion along the x axis (see viewtopic.php?f=3&t=261&start=10#p797).
However, since my file had (x,y) experimental data, the fit was correct. The display on the fitting window was distorted, but not the actual fit. However, I was using
Maybe you should just convert your file to (x,y) and just ignore the visual distortion during fitting?
However, since my file had (x,y) experimental data, the fit was correct. The display on the fitting window was distorted, but not the actual fit. However, I was using
curry, not pepper, and (T,chi*T) data, so there might be a difference there.Maybe you should just convert your file to (x,y) and just ignore the visual distortion during fitting?