Hi,
I’m trying to implement a data-driven estimation of the QCD fake-leptons background for top quark physics based on the so-called “matrix method” using HistFitter. Following the examples, I see I can estimate the signal strength and sample scale factors based on pre-made histograms, but in my case I would like to find a way to build the background template as a function of other samples (namely, data, signal and other MC background). Is there a neat way to do it in HistFitter? In particular, I need to estimate the fake and real efficiencies for tight and loose selections, which are in turn yet other parameters to be evaluated.
Any suggestion would be very appreciated.
Cheers,
Riccardo