July 31, 2017 to August 4, 2017
Fermi National Accelerator Laboratory
US/Central timezone

Histogram Binning with Bayesian Blocks

Aug 3, 2017, 11:29 AM
Hornets Nest (Fermi National Accelerator Laboratory)

Hornets Nest

Fermi National Accelerator Laboratory

Presentation Computing, Analysis Tools and Data Handling Computing, Analysis Tools, and Data Handling


Dr Brian Pollack (Northwestern University)


The Bayesian Block algorithm, originally developed for applications in astronomy, can be used to improve the binning of histograms in high energy physics. Along with visual improvements, the histogram produced from this algorithm is a non-parametric density estimate, providing a description of background distributions that do not suffer from the arbitrariness of ad-hoc analytical functions. The statistical power of a hypothesis test based on a Bayesian Blocks binned template is nearly as good as one obtained by fitting analytical functions. This presentation will showcase the visual and statistical benefits of the Bayesian Blocks algorithm on a handful of examples based on common HEP analyses.

Primary author

Dr Brian Pollack (Northwestern University)


Michael Schmitt (Northwestern University) Ms Saptaparna Bhattacharya (Brown University)

Presentation materials