Hi,
I am a graduate physics student and am somewhat new to ROOT and I have been trying to fit a spectrum from our ADC in our data acquisition system. The spectrum is from a PMT looking at a small scintillator with an Americium241 source attached to it. It looks somewhat like a Landau distribution but trying such a fit from root has been giving sub par results. See below.
The code I used for doing this fit looked like:
TF1 *americium = new TF1(“americium”,"([0]*TMath::Landau(x,[1],[2])) + [3]", 110., 350.);
americium->SetParameter(0,500.);
americium->SetParameter(1,165.);
americium->SetParameter(2,50.);
americium->SetParameter(3,50.);
I also had postdoc in my lab take a look and he tried a fit that looked like a Gaussian times a Landau which gives great results but I just don’t understand them. The point of the fitting is to properly extract the peak or most probable value of the distribution and the parameters for this fit don’t seem to correlate with where the peak is (looks be around 166 ADC). In fact the first fit gives a closer value even though it looks terrible.
The code for this alternate fit looks like:
TF1 *americium = new TF1(“americium”,"([4]*TMath::Gaus(x,[0],[1]))*TMath::Landau(x,[2],[3])", 110., 350.);
americium->SetParameter(0,165);
americium->SetParameter(1,50.);
americium->SetParameter(2,165.);
americium->SetParameter(3,50.);
americium->SetParameter(4,500.);
As you can see in the plot the fit is very good but the parameters root spits out diverge wildly from what I would expect and don’t seem to tell me much at all about where the peak is (which is the whole goal).
Any advice on how to go about fitting this peak would be greatly appreciated.
Thanks,
John