TSpectrum class

ROOT
1

Hi,

I just spotted the TSpectrum class ( root.cern.ch/doc/master/classTSpectrum.html , didn’t know it up to now). Looks fine but I’m not quite sure if I can use it for my purposes since the background in all its examples is obviously very different from the signals.
In my case I have only two peaks: One noise (the left) and one signal peak and mostly in between they are overlaid (also besides).
So my question is: Does it make sense to use this class for my case ? Otherwise I would try to find a fit which concerns the noise peak (the signal peak follows a landau distribution but I need something else because the noise peak moves the landau fit).

Thank you very much in advance!

Selection_120.png889×660 3.59 KB

2

Hi,

I think the TSpectrum class is not really for your case. I would just try fitting your data with a sum of gaussian or something similar more sophisticated

Best Regards

Lorenzo

3

Hi,

thank you! A colleague said quite the same. Because of curiousity: Why do you think it won’t help ? The examples of the deconvolution method look a kind of similar. I just could try it but I have much more to adjust in the existing code than the small block promises :slight_smile:

Kind regards,
Ben