Scaling signal rate back to its original value in pull distributions in bias studies

Roofit and RooStats
1

Hi all,

I’m doing bias studies using the CMS Higgs Combine Tool where I calculate the pull distribution defined as (mu_fit - mu_true)/mu_err. Because of my extremely low signal rate, i.e., at an order of (x1e-11), I’m having fit problems. However, if I scale my signal rate to (x1e+11), my fit turns fine. This so far is the best solution for me to perform bias studies with the cms higgs combine tool (combine tool puts the mu in the O(1) range), that is, rescale my signal rate to make the fits fine. Now, how do I re-scale my signal rate back to its original value and hence, obtain correct pull distributions?

Thanks a lot!

2

No idea. Is there a list / forum of combine experts in CMS?

3

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.