Fitting a double-peak distribution using Root

Roofit and RooStats
1

I have a distribution that has a double-peaked structure. I am trying to fit this using a parabola exhibiting a minimum convolved with a Gaussian resolution function. Can I use RooFFTConvPdf to do this? Or is there a way to define the convolution function (which will be an integration function) in ROOT and do the fitting? Which would be the best approach? Any other pdfs that can describe the given shape is also welcome. I really appreciate any help you can provide.

eqn.pdf (6.8 KB)
dE_Dpi0_signal.pdf (14.0 KB)

2

Hi @Ansu,

First of all, welcome to the ROOT forum! Regarding your issue, maybe @moneta or @jonas can help here.

Cheers,
J.

3

Hi,

You can use the RooFFTConvPdf if your pdf you are convoluting is going to zero at the border of your range, otherwise you will have circular issues by usingFFT’s. Otherwise you can always use numerical convolution (class RooNumConvPdf) based on computing the integral directly, which will be slower.
@jonas can help you in case you have any RooFit issues with this

Best regards

Lorenzo

4

@moneta ,
Thank you so much for the reply. So in this case, since I am trying to convolve a parabola that does not go to zero at the border, is the best option RooNumConvPdf?

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.

6

Yes in this case it is better to use the RooNumConvPdf.