Sixth exercise (EXOMJET)

  • Goals of Exercises:
    • Red led STEP 1: Produce AOD single-jet skim from EXOSingleJet sample (or SD_Jet50U if not available) and store in an EXO T2. See details here.
    • Red led STEP 2: Produce AOD single-muon skim from EXOMu sample (or SD_Mu9 if not available) and store in an EXO T2. This step has been not tried from the subgroup. The full workflow has been extensively trained by other EXO subgroup.
    • Red led STEP 3: Produce root-tuples (PAT) for analysis via CRAB . See details here.
    • Red led STEP 4: Produce histograms (basic MHT spectra and angular distributions) and efficiency tables of analysis results. See details here.

  • Contact: Leonardo Benucci
  • Analysis Group Members: LB, Marco Cardaci, Thomas Maes
  • T2s involved: GRIF, RALPP, Purdue, JINR


We ran the exercise on the following samples:
  • Secondary dataset SD_Jet50U (RECO). The following samples for 10 and 7 TeV have been used:
    • /QCD_Pt30/Summer09-MC_31X_V3_SD_Jet50U-v1/GEN-SIM-RECO (13.8GB)
    • /QCD_Pt80/Summer09-MC_31X_V3_SD_Jet50U-v1/GEN-SIM-RECO (624.2GB)
    • /QCD_Pt30/Summer09-MC_31X_V3_7TeV_SD_Jet50U-v1/GEN-SIM-RECO (29.5GB)
    • /QCD_Pt80/Summer09-MC_31X_V3_7TeV_SD_Jet50U-v1/GEN-SIM-RECO (812.0GB)
  • Group skim EXOSingleJet, where jet pass basic quality cuts and the leading jet is required to have pT> 100 GeV. It has been produced by priority user on GRIF T2. The following samples for 10 and 7 TeV have been used:
  • /QCD_Pt30/exotica-EXOSingleJetSkimOct09_7TeV-2c4dd3844f8fd004251ec676ec977768/USER from a local DB
  • /QCD_Pt80/exotica-EXOSingleJetSkimOct09-2c4dd3844f8fd004251ec676ec977768/USER from global DB.


We used a PAT-tuple producer for SUSY in Most of Grid jobs went smoothly with no major bottlenecks ad 100% success fraction. Only in the EXOSingleJetSkimOct09_7TeV the rate was about 70%.
The output has been dumped directly on Castor.


A root-tree containing basic kinematical quantities for jets and MET have been produced with a private code. Four different distributions on QCD samples have been produced:

  • raw MET, as available in PAT
  • pT of all jets with pT>20 GeV and |eta|<3
  • multiplicity of the jets above
  • MHT, performed off-line with all jets with pT>20 GeV and |eta|<5.
These distribution has been defined following what was used in mono-jet analysis documented in PAS EXO-09-013 about mono-jet analysis. A comparison among CaloJet, PFJet and Jet+Track and among CaloMET, PFMET and tCMET has been carried on.

The pictures below are an example of results for SD_Jet50U and for EXOSingleJet, both for 7 TeV.

Some distributions of jet and MET quantities for results for SD_Jet50U (7 TeV).
Some distributions of jet and MET quantities for results for EXOSingleJet (7 TeV).

From the comparison of two sets, it can be understood that:

  • pT distribution is consistent with the fact that the content of hard jet is enhanced
  • MET and MHT distributions are consistent, since most of low-energy jet have been removed in skim
  • the bulk of jet multiplicity distribution is consistent between the two skims. Significant depletion of low-multiplicity jets after the skim is due to events with a single low-pT jet not passing the cuts
From the comparison of different objects, it can be understood that:
  • as expected, PF calculation of MET has a lower mean value of tcMET, that is in turn lower than caloMET
  • usage of MHT largely reduces the tcMET and caloMET discrepancy, with the replacement of energy cluster with track pT becoming important only for MHT>100 GeV
  • PF and JPT jet definition are effective in supressing the jets in high-multiplicity events and produce a maximum at two jets as expected


These simple comparisons lead to the conclusion that PFMET/tcMET and and PFjet/JPT are better reproducing the kinematic of QCD events. In particular, it is important to use these flavours when cutting on MET, MHT, pT and jet multiplicity. If MET is not considered, usage of PF object can be replaced by Jet+Track

-- LeonardoBenucci - 16-Nov-2009

Topic attachments
I Attachment History Action Size Date Who Comment
GIFgif EXOMJ_80.gif r1 manage 17.0 K 2009-11-17 - 11:12 LeonardoBenucci  
GIFgif SD_QCD80.gif r1 manage 17.0 K 2009-11-17 - 11:12 LeonardoBenucci  
Edit | Attach | Watch | Print version | History: r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r3 - 2009-11-18 - LeonardoBenucci
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Sandbox All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback