I just tried a simple linear fit, but with the instrumental weighting method (means: I added specific error bars to the graph and fitted it with these errors). It didn't work out so well (see below) and I'm absolutly clueless right now. Would be really nice, if someone knows what to do and can help me.
Tobi
[09.05.2019 15:03:30 Plot: ''R=f(U)'']
Non-linear fit of dataset: Tabelle1_I[mA], using function: a*x
Y standard errors: Associated dataset (1)
Scaled Levenberg-Marquardt algorithm with tolerance = 0,0001
From x = 0 to x = 19,96
a = 1 +/- nan
Chi^2 = nan
R^2 = nan
Iterations = 0
Status = iteration is not making progress towards solution
There are two possible workarounds for you:
1- Remove the first line of your table (the line which contains zeros), and try to fit using the y=ax function (works here);
2- Instead of zero, in the first cell of column 3 (the yerr column), put there a very small value, say 1e-50, and perform a linear fit with the function y=ax+b (works here, but y=ax do not).
Best regards.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I just tried a simple linear fit, but with the instrumental weighting method (means: I added specific error bars to the graph and fitted it with these errors). It didn't work out so well (see below) and I'm absolutly clueless right now. Would be really nice, if someone knows what to do and can help me.
Tobi
[09.05.2019 15:03:30 Plot: ''R=f(U)'']
Non-linear fit of dataset: Tabelle1_I[mA], using function: a*x
Y standard errors: Associated dataset (1)
Scaled Levenberg-Marquardt algorithm with tolerance = 0,0001
From x = 0 to x = 19,96
a = 1 +/- nan
Chi^2 = nan
R^2 = nan
Iterations = 0
Status = iteration is not making progress towards solution
Last edit: Tobias T 2019-05-09
Hi Tobias!
There is a bug here. Not in SciDAVis, but in GSL (the routine used to perform the fit). See: https://sourceforge.net/p/scidavis/scidavis-bugs/277/#6ca8
There are two possible workarounds for you:
1- Remove the first line of your table (the line which contains zeros), and try to fit using the y=ax function (works here);
2- Instead of zero, in the first cell of column 3 (the yerr column), put there a very small value, say 1e-50, and perform a linear fit with the function y=ax+b (works here, but y=ax do not).
Best regards.