Hi,
I am still quite new to ROOT. I don’t know if it is possible to calculate the signal significances before and after applying some cuts. If it’s possible can you give some basic ideas and/or references that I may look at to learn this?
It will be a great help to me.
Unfortunately your question is most likely not as simple as it sounds. If you perform a counting experiment, you can use the ROOT function here, which also supports an uncertainty on the background. You can read the details in the paper here. If you have a more complicated setup for your analysis, the book here is a good resource.
Thank you so much, Stefan, for the quick answer. And I am sorry for such a delay.
I will be using Delphes output files for the signal as well as the background. I looked at the references suggested by you. But I saw that I need PDFs. Now, how do I get the PDF of some histogram(s)? I mean is it possible?
So far what I was doing is using the metric “S/sqrt(S+B)”, with S and B are event number obtained by “cross-section * Luminosity” I was calculating the significance. I suppose this is wrong, right?
But from the references what I found is using a particular histogram the signal significance is being calculated. So what happens if I take different histograms of the same analysis?
I am sorry if this is a very disturbing question. I am new to this field so I don’t have that much understanding yet.
In this forum, we discuss the technical aspects of ROOT, but unfortunately, we don’t have time to discuss topics that you can find in textbooks. I would recommend you contact supervisors/colleagues or look for nice textbooks on statistics.