Rebalance and Smear for Multi-jet Background Estimation

TL;DR

This paper presents the Rebalance and Smear (R&S) method for estimating multi-jet backgrounds in collider searches, providing a detailed description, software implementation, and a case study on dark matter searches.

Contribution

It offers a comprehensive top-to-bottom explanation of the R&S method, including its applicability, limitations, and a software tool interfacing with Delphes3 data for background prediction.

Findings

The R&S method effectively predicts QCD multi-jet backgrounds.

The software implementation facilitates practical application in collider analyses.

Systematic uncertainties and potential biases are analyzed and addressed.

Abstract

For many particle collider searches for physics beyond the Standard Model in final states with jets and missing transverse momentum, events from QCD multi-jet processes are an important and challenging background contribution. The CMS and ATLAS experiments have previously developed data-driven methods designed to succeed where Monte Carlo methods suffer large theoretical and experimental uncertainties. One such method is Rebalance and Smear (R&S), which predicts QCD backgrounds by applying a series of folding and unfolding techniques to data control regions. A top-to-bottom description of the R&S method is presented, along with a discussion of its applicability and limitations. A software application is provided that performs the R\&S method using public, non-proprietary tools, interfacing with data sets produced by Delphes3. In general, the method is suitable for predicting multi-jet backgrounds in searches for stable non-detectable particles, such as dark matter candidates. A case study is carried out in simulated events of proton-proton collisions at 14 TeV in the context of a potential search for Higgsino dark matter produced in the decay products of supersymmetric quark partners. Sources of potential bias are explored and prescriptions for evaluating systematic uncertainties are suggested.