Pileup Jet Identification

PAS : JME-13-005

Abstract

High pileup in LHC collisions can increase incidence of jets by several large factors. To reduce the incidence of jets from pileup and to preserve the rate of good jets, a jet identification based on both vertex information and jet shape information has been developed.The construction of this jet identifier is described and the performances are evaluated using both Z+jets MC simulated samples and Z+jets data collected in the 2012 $\sqrt{s}=8$~TeV run. The effectiveness of this jet identifier is discussed in the context of jet vetoes and vector boson fusion production.|

Plots from the paper JME-13-005 ( click on plot to get .pdf )

Toy MC Plots

Rate of pileup jets computed by considering the probability overlapping pileup jets of lower pT, which sum up to a larger single jet of sum pT given by the individual components. The individual jet probabilities are determined by the jet cone size compared to the CMS jet volume using a predicted inclusive jet spectrum per collisions of 1/pT^5.

JME-13-005 Plots

	Rate of data and MC PU jets with pt>25GeV relative to the expected rate of real jets as a function of the number of reconstructed primary vertices.
	Jet pT distribution for all jets having a pT > 25GeV for the full 2012 dataset.
	Jet eta distribution for all jets having a pT > 25 GeV for the full 2012 dataset.
	Comparison between jet flavors and pileup for jets with pT > 25GeV for the track related variable: beta
	Comparison between jet flavors and pileup for jets with pT > 25GeV for the track related variable: betaStar
	Comparison between jet flavors and pileup for jets with pT > 25GeV for the track related variable: dZ
	Comparison between jet flavors and pileup for jets with pT > 25GeV for the track related variable: number of primary vertices
	$\langle\Delta R^2 \rangle$ for PF jets with pT > 25GeV and eta < 2.5
	$\langle\Delta R^2 \rangle$ for PF jets with pT > 25GeV and eta > 3.0
	A < (\Delta R) < A+0.1 for PF jets with pT > 25GeV and eta < 2.5 for A=0.1
	A < (\Delta R) < A+0.1 for PF jets with pT > 25GeV and eta < 2.5 for A=0.2
	A < (\Delta R) < A+0.1 for PF jets with pT > 25GeV and eta < 2.5 for A=0.3
	A < (\Delta R) < A+0.1 for PF jets with pT > 25GeV and eta < 2.5 for A=0.4
	A < (\Delta R) < A+0.1 for PF jets with pT > 25GeV and eta < 2.5 for A=0.5
	Number of charged particles for PF jets with pT > 25GeV and eta < 2.5
	Number of neutral particles for PF jets with pT > 25GeV andeta < 2.5.
	$p_{T}^{D}$ for PF jets with pT > 25GeV and eta < 2.5.
	MVA discriminator for particle flow jets with p_{T} > 25GeV and eta < 2.5.
	MVA discriminator for particle flow jets with p_{T} > 25GeV and 2.5 < eta < 2.75. Disagreement in the pileup region of the MVA is present in the region where 2.5<eta. This is a known effect, which results from improper simulation of out-of-time pileup.
	MVA discriminator for particle flow jets with p_{T} > 25GeV and 2.75 < eta < 3.0. Disagreement in the pileup region of the MVA is present in the region where 2.5<eta. This is a known effect, which results from improper simulation of out-of-time pileup.
	MVA discriminator for particle flow jets with p_{T} > 25~\GeV and 3.0 < eta < 5.0. Disagreement in the pileup region of the MVA is present in the region where 2.5<eta. This is a known effect, which results from improper simulation of out-of-time pileup.
	ROC curves for quark and gluon jets with 25 < pT < 100gev for eta < 2.5
	ROC curves for quark and gluon jets with 25 < pT< 100gev for 2.5 < eta < 2.75
	ROC curves for quark and gluon jets with 25 < pT < 100 gev for 2.75 < eta < 3.0
	ROC curves for quark and gluon jets with 25 < pT < 100 gev for 3.0 < eta < 5.0
	Data-MC comparison of the MVA (loose working point) pileup jet identification efficiency versus eta on the Z to mu mu+Jets sample for PF jets with pT > 25GeV
	Data-MC comparison of the MVA (loose working point) pileup jet identification efficiency versus the number of primary vertices on the Z to mumu +jets sample for PF jets with pT > 25~GeV
	Data-MC comparison of the MVA (loose working point) pileup jet identification efficiency versus jet pt on the Z(to mumu )+jets sample for PF jets with pT > 25GeV

Usage Examples

	Rate of migration of 0 jets to 1 jet events for a jet with pT above the threshold given by the axis for Z to dimuon events before the pileup jet is applied.
	Rate of migration of 0 jets to 1 jet events for a jet with pT above the threshold given by the axis for Z to dimuon events after the pileup jet is applied.
	Rate of all jets to real jets before and after the pileup jet id is applied for jets with a pT cut greater 25 GeV and eta < 2.5
	Rate of all jets to real jets before and after the pileup jet id is applied for jets with a pT cut greater 25 GeV and eta > 2.5
	Rate of all jets to real jets before and after the pileup jet id is applied d for jets passing a generic VBF selection having two jets with a pT cut greater than 30 GeV , the mass of the dijet system is greater than 500 GeV and the difference in eta between the two jets is > 3.5