Limits, discovery and cut optimization for a Poisson process with uncertainty in background and signal efficiency: TRolke 2.0

TL;DR

TRolke 2.0 is a C++ package that calculates confidence intervals and hypothesis tests for Poisson processes with various uncertainties, aiding in precise statistical analysis in physics experiments.

Contribution

It introduces a versatile C++ class supporting multiple uncertainty combinations for confidence interval calculations and hypothesis testing in Poisson processes.

Findings

Supports seven uncertainty combinations including Binomial, Gaussian, and Poisson.

Provides routines for upper/lower limits and sensitivity calculations.

Compatible with C++, Python, and ROOT for flexible usage.

Abstract

A C++ class was written for the calculation of frequentist confidence intervals using the profile likelihood method. Seven combinations of Binomial, Gaussian, Poissonian and Binomial uncertainties are implemented. The package provides routines for the calculation of upper and lower limits, sensitivity and related properties. It also supports hypothesis tests which take uncertainties into account. It can be used in compiled C++ code, in Python or interactively via the ROOT analysis framework.