Iterative unfolding with the Richardson-Lucy algorithm

TL;DR

This paper presents an improved iterative unfolding method using the Richardson-Lucy algorithm, including an optimal iteration fixing procedure, analysis of initial conditions, and a documentation approach for comparison with theoretical models.

Contribution

It introduces a new algorithm to determine the optimal number of iterations in Richardson-Lucy unfolding and proposes a standardized way to document results for comparison.

Findings

Optimal iteration fixing algorithm tested on various distributions

Starting distribution influences unfolding results

Method enables comparison with theoretical predictions

Abstract

The Richardson-Lucy unfolding approach is simple and excellently performing. It efficiently suppresses artificial high frequency contributions and permits to introduce known features of the true distribution. An algorithm to fix the number of iterations to an optimal value has been developed and tested with five different types of distributions, with different event numbers and with different binnings. The influence of the starting distribution has been studied. A simple way to document the unfolding result such that it can be compared to theoretical predictions is proposed.