Tools for corrections (AtlfastCorrectors)

Introduction

The purpose of this package is to apply lepton efficiency and fake rate to the outputs of Atlfast. The main features are as follows.

  • lepton efficiency and fake rate are applied as a function of pT and eta.
  • scale shift by the misidentification is supported.
  • fake missing ET from muon inefficiency and jet smearing is taken into account (recalculating technique instead of reweighting).
  • fake leptons from tau hadronic jets are separately treated.
  • transfer functions for jets and missing eT are tuned with full sim data.

The series of Algorithms make the corrected version of AOD containers from the original containers. Therefore you can use the corrected outputs by just switching names of the containers in your analysis code or processing script (EventView etc). If you want to change the default parameters, you can put the parameters in any format (values, function, histogram). Futhermore, the full chain of the corrections are managed in job options. It means you can customize the correction chain, how to apply fake, efficiency, overlap removal or how to deal with photon, tau. There are some examples in share (AtlfastCorrectionAlgs_TestXXXX.py).
contact: Yousuke.Kataoka@cernNOSPAMPLEASE.ch

Document

Usage

At first, install on your test area. The latest source codes are here
source setup.sh
cd $TestArea
mkdir PhysicsAnalysis
cd PhysicsAnalysis
tar xvfz AtlfastCorrectors.tar.gz
cd AtlfastCorrectors/cmt
cmt broadcast cmt config
source setup.sh
cmt broadcast make
Then, insert the following statement before your Algorithms in your job option.
include ( "AtlfastCorrectionAlgs/AtlfastCorrectionAlgs_CorrectionsForSusy.py" )
And you can use corrected version of AOD containers with the following names
  • CorrectedAtlfastElectronCollection
  • CorrectedAtlfastMuonCollection
  • CorrectedAtlfastNonIsoMuonCollection
  • CombinedJets -> CombinedParticleJet before Jet migration
  • CombinedMissingEt
The default parameters for leptons are tuned for SUSY WG standard Lepton ID(e: egamma,medium,mu:STACO,combined,match chi2<100) with full sim samples. The overlaps between jets and photons(or taus) are not removed. The another jet containers are prepared for special uses
  • CombinedJetsIncMuon (including pT of muon in jet) -> CombinedParticleJetsIncMuon before Jet migration
  • TransferedCone4TruthJets (overlaps with electrons are not removed, and not includeing muon) -> TransferedCone4TruthParticleJets before Jet migration
The tau containers are also made
  • TransferedAtlfastTauJetContainer
  • TransferedAtlfastTauJet1p3pContainer
in the above containers, the same tau IDs with default Atlfast are applied on top of transfered jets.

Also you have b-tagged container

  • TransferedAtlfastBJetContainer

If you want to dump corrected containers to the persistencies from AOD file, look at the following examples in share

  • AtlfastCorrectionAlgs_AODtoTFtoAOD.py
  • AtlfastCorrectionAlgs_AODtoTFtoAAN.py
  • AtlfastCorrectionAlgs_AODtoTFtoSV.py (for rel.12)

known problem

run on Atlfast-I sample (McLocation="GEN_EVENT")

The name of MC history is "GEN_EVENT" in Atlfast-I or Evgen sample and not "GEN_AOD" in AOD sample. If you run on Atlfast-I sample, add the following line before including the main job option
McLocation = "GEN_EVENT"
include ( "AtlfastCorrectionAlgs/AtlfastCorrectionAlgs_CorrectionsForSusy.py" )

Corrections for SUSY WG

atlfcorr2.gif

Electron efficiency

Efficiency is applied as a function of pT and eta. The functions are in the form of histograms which are derived from T1 sample. Inefficient electrons appear as jets. The above plot is for rel.12, the efficiency is updated by rel.14 in the release

Muon efficiency

Efficiency is applied as a function of pT and eta. The functions are in the form of histograms which are derived from T1 sample. Inefficient muons are considered as missing ET. The above plot is for rel.12, the efficiency is updated by rel.14 in the release

Electron fake rate

Fake rate is provided as a function of pT. The function are in the form of histograms which are derived from T1 sample. Fake electron from tau jet and fake electron from non-tau jet are separately treated since the difference of the fake rate is about a factor of 10. Scale shift in misidentification is given as a probability function, which is in the form of histograms and provided for various pT regions of non-tau jet and tau jet. The above plot is for rel.12, the efficiency is updated by rel.14 in the release

Muon fake rate

Fake rate for muon is less than 1/10 of that for electron. Therefore it is neglected for simplicity.

Photon

Photons are not reconstructed and appear as jets.

Missing ET

jet energy, soft component and muon inefficiency are added as fake missing ET.

Validation

Electron efficiency

  • comparison between full sim. and corrected Atlfast for pT dependence of electron efficiency (left figure) ... valElectronEfficiencyPt.gif
  • comparison between full sim. and corrected Atlfast for eta dependence of electron efficiency (right figure) ... valElectronEfficiencyEta.gif
    valElectronEfficiencyPt.gif valElectronEfficiencyEta.gif
The above plot is for rel.12, the efficiency is updated by rel.14 in the release

Muon efficiency

  • comparison between full sim. and corrected Atlfast for pT dependence of muon efficiency (left figure) ... valMuonEfficiencyPt.gif
  • comparison between full sim. and corrected Atlfast for eta dependence of muon efficiency (right figure) ... valMuonEfficiencyEta.gif
    valMuonEfficiencyPt.gif valMuonEfficiencyEta.gif
The above plot is for rel.12, the efficiency is updated by rel.14 in the release

Electron fake rate

  • comparison between full sim. and corrected Atlfast for electron fake rate from non-tau jet (left figure) ... valElectronFromNonTau.gif
  • comparison between full sim. and corrected Atlfast for electron fake rate from tau jet (right figure) ... valElectronFromTau.gif
    valElectronFromNonTau.gif valElectronFromTau.gif

  • comparison between full sim. and corrected Atlfast for pT distribution of fake electron ... valFakeElectronPt.gif
    valFakeElectronPt.gif
The above plot is for rel.12, the fake rate is updated by rel.14 in the release

Fake missing ET

  • comparison between full sim. and corrected Atlfast for fake missingET (T1) ... t1.gif
    t1.gif
(for events without muon>2.5)

  • comparison between full sim. and corrected Atlfast for fake missingET (J3:left, J6:right) ... j3.gif, j6.gif
    j3.gifj6.gif

reference


Lepton efficiency, fake rate comparison within SUSY WG

The page is dedicated for numbers for lepton efficiency, fake rate comparison between different analysis groups within the SUSY WG. The page includes discussion, slides, results within SUSY WG and also with the egamma and other working group.

The first meeting was on 7th June 2007 to discuss about the possible agenda need to be followed for these comparison. The agenda page can be found in here.

Slide from different groups ( 18th July 2007 )

Talks given in meetings

Should follow Samples and Dataset

T1 : [ trig1_misal1_mc12.005200.T1_McAtNlo_Jimmy.recon.AOD.v12000604_tid008037 ]

SU3 : [ trig1_misal1_csc11.005403.SU3_jimmy_susy.recon.AOD.v12000601_tid006978 ]


Topic attachments
I Attachment History Action Size Date Who Comment
Unknown file formatgz AtlfastCorrectors.rel13.tar.gz r12 r11 r10 r9 r8 manage 61.2 K 2009-05-08 - 17:10 YousukeKataoka AtlfastCorrectors (ver. 00-01-05) for release 13
Unknown file formatgz AtlfastCorrectors.rel14.pj.tar.gz r4 r3 r2 r1 manage 61.3 K 2009-05-08 - 17:14 YousukeKataoka AtlfastCorrectors (ver. 00-02-05) for release 14, ParticleJet
Unknown file formatgz AtlfastCorrectors.rel14.tar.gz r17 r16 r15 r14 r13 manage 195.6 K 2009-11-08 - 16:37 YousukeKataoka AtlfastCorrectors (ver. 00-02-08) for release 14
Unknown file formatgz AtlfastCorrectors.rel15.tar.gz r2 r1 manage 194.4 K 2009-11-08 - 16:37 YousukeKataoka AtlfastCorrectors for release 15
Unknown file formatgz AtlfastCorrectors.tar.gz r41 r40 r39 r38 r37 manage 86.5 K 2009-11-06 - 10:54 YousukeKataoka AtlfastCorrectors (ver. 00-00-13) for release 12
PDFpdf FrankPaige_13.0.10_Validation.pdf r1 manage 1286.7 K 2007-07-23 - 22:41 TapasSarangi  
PowerPointppt Muon_performances_ATLAS_week.ppt r1 manage 95.5 K 2007-10-09 - 21:57 IrisBorjanovic  
PDFpdf egamma_csc1_170707.pdf r1 manage 154.0 K 2007-07-23 - 13:58 TapasSarangi  
GIFgif j3.gif r2 r1 manage 9.5 K 2008-05-18 - 09:08 YousukeKataoka fake missing ET for J3
GIFgif j6.gif r2 r1 manage 9.4 K 2008-05-18 - 09:07 YousukeKataoka fake missing ET for J6
PowerPointppt muons_software_week.ppt r1 manage 481.0 K 2007-10-24 - 15:41 IrisBorjanovic update of slides from Lecce/Napoli
GIFgif t1.gif r2 r1 manage 9.1 K 2008-05-18 - 09:07 YousukeKataoka fake missing ET for T1 (tt sample)
Edit | Attach | Watch | Print version | History: r67 < r66 < r65 < r64 < r63 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r67 - 2009-11-08 - YousukeKataoka
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback