With the option smoot acspline I can get very nice and well aproximation of my data. My question is, however, is it possible to know the final spline fitting coefficients? For example, the interception point with y axis and its error.
I have an additional question. After a fitting of a data file with some predefined function I can print to the terminal the function fitting coefficients and their errors. How can I print the rest of the fitting statistics like the rms of the fit etc.? Of course, I can find this information in the fit.log file, but if they are stored as some kind of variables, I would like to know how can I print them.
Thanks,
iskren
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
'smooth' is not meant to approximate data by a function in any useful way, nor is it really any good at it. The reason you don't get the spline parameters back, but only the actual curve is that only the curve is meaningful.
As for the other request, most of that has just recently been added to gnuplot as a new feature.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hellow!
With the option smoot acspline I can get very nice and well aproximation of my data. My question is, however, is it possible to know the final spline fitting coefficients? For example, the interception point with y axis and its error.
I have an additional question. After a fitting of a data file with some predefined function I can print to the terminal the function fitting coefficients and their errors. How can I print the rest of the fitting statistics like the rms of the fit etc.? Of course, I can find this information in the fit.log file, but if they are stored as some kind of variables, I would like to know how can I print them.
Thanks,
iskren
'smooth' is not meant to approximate data by a function in any useful way, nor is it really any good at it. The reason you don't get the spline parameters back, but only the actual curve is that only the curve is meaningful.
As for the other request, most of that has just recently been added to gnuplot as a new feature.