At first I have one dataset and calculate the signal yield n0 by fitting one distribution in roofit. Now I have another data set with some conditions and find out the signal yield n1 after fitting the same distribution.
I want to calculate the efficiency = n1/n0 bin by bin. Is there any way to calculate the efficiency like a distribution after getting the yield from fit. Many thanks in advance.
I don’t understand. The signal yields n1 and n0 are obtained from a fit and are not a bin/bin distribution.
I guess you will need to do a simultaneous fit to the ratio of the distribution, but I need more information and a clear description of your data and model to understand the problem better
Thank You. I am reconstructing j/psi from two muons without any particle identification on muon. I fit j/psi mass and calculate the signal and background yield. Later I applied particle identification condition on the muon and fit j/psi mass and calculate signal and background yield.
Now I am trying to calculate , what is the efficiency of the particlr identification, bin by bin.
What is the variable you are using to bin the efficiency ? It cannot be the di-muons invariant mass because you will not be sensible outside the H./psi peak
Currently, I am taking invariant mass for calculating the identification efficiency. Later I may consider some other variables. I think, TBinomialEfficiencyFitter will be useful for me.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.