Alternative angular variables for suppression of QCD multijet events in new physics searches with missing transverse momentum at the LHC

TL;DR

This paper introduces three new angular variables to better suppress QCD multijet background events in LHC supersymmetry searches involving missing transverse momentum, improving upon existing methods.

Contribution

The paper proposes three novel angular variables, $ ilde{ ext{omega}}_ ext{min}$, $ ext{hat} ext{omega}_ ext{min}$, and $ ext{chi}_ ext{min}$, that outperform previous variables in rejecting background events in high jet multiplicity scenarios.

Findings

$ ext{hat} ext{omega}_ ext{min}$ and $ ext{chi}_ ext{min}$ outperform $ ext{Delta} ext{phi}^*_ ext{min}$ and $ ext{Delta} ext{phi}_i$ in background rejection.

The new variables effectively reduce total standard model background events.

Simulated data demonstrates improved background suppression with the new variables.

Abstract

We introduce three alternative angular variables-denoted by $\tilde{\omega}_\text{min}$, $\hat{\omega}_\text{min}$, and $\chi_\text{min}$-for QCD multijet event suppression in supersymmetry searches in events with large missing transverse momentum in proton-proton collisions at the LHC at CERN. In searches in all-hadronic final states in the CMS and ATLAS experiments, the angle $\Delta\varphi_i$, the azimuthal angle between a jet and the missing transverse momentum, is widely used to reduce QCD multijet background events with large missing transverse momentum, which is primarily caused by a jet momentum mismeasurement or neutrinos in hadron decays-the missing transverse momentum is aligned with a jet. A related angular variable-denoted by $\Delta\varphi^*_\text{min}$, the minimum of the azimuthal angles between a jet and the transverse momentum imbalance of the other jets in the event-is used instead in a series of searches in all-hadronic final states in CMS to suppress QCD multijet background events to a negligible level. In this paper, before introducing the alternative variables, we review the variable $\Delta\varphi^*_\text{min}$ in detail and identify room for improvement, in particular, to maintain good acceptances for signal models with high jet multiplicity final states. Furthermore, we demonstrate with simulated event samples that $\hat{\omega}_\text{min}$ and $\chi_\text{min}$ considerably outperform $\Delta\varphi^*_\text{min}$ and $\Delta\varphi_i$ in rejecting QCD multijet background events and that $\hat{\omega}_\text{min}$ and $\tilde{\omega}_\text{min}$ are also useful for reducing the total standard model background events.