Question about Beeston Barlow Method in histFactory: Handling Weighted Templates

Roofit and RooStats
1

Dear experts,

I have been trying to understand if the Beeston Barlow light method implemented in histFactory can handle weighted templates, i.e., are the uncertainties reported by a fit based on weighted templates accurate? My question has been triggered by the latest PHYSTAT seminar “Template fits - Fitting non-parametric density models to data” by Dr. Hans Dembinski (see: Template fits: fitting non-parametric density models to data (10 April 2024) · Indico). From this talk, I can understand that an implementation using the Beeston Barlow method with Conway’s approximation and SPD should be able to handle weighted templates; however, from the histFactory documentation (there: https://cds.cern.ch/record/1456844/files/CERN-OPEN-2012-016.pdf), it is not clear to me that this is used by histFactory.

Thank you for the help.

Cheers,

Veronica

2

Hi Veronica,

Thanks for this very abstract but interesting post.
In case he has pointers, I add in the loop @jonas .

Cheers,
Danilo