I am interested in how ROOTFIT implement RooHistPdf to transfer a histogram data ( are they using histogram or data?) to PDF and how they use generation method to genrate toy model data. Can anyone please tell me any introduction of the theory?
RooFit uses a binned axis, and fills the data into these bins. From this, you can construct a binned PDF by dividing each bin by the total event count.
I don’t know how generation is implemented, but it should be simple to do generation by constructing the CDF over all bins, throwing uniformly distributed random numbers, and inverting the CDF at the location suggested by the random numbers.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.