wiki:Korea2017-08-ATL-PHYS-PROC-2016-117

Version 23 (modified by Eric CONTE, 7 years ago) ( diff )

--

The first MadAnalysis 5 workshop on LHC recasting @ Korea

Analysis 8: ATL-PHYS-PROC-2O16-117

Students
Supervisor(s) Eric Conte Benjamin Fuks

1. Analysis summary

A measurement of the tt¯Z and tt¯W production cross sections in final states with:

  • two same-charge muons: one muon coming from Z/W and one from the top decay.
  • or three leptons (electrons or muons),
  • or four leptons (electrons or muons): only ttZ case where Z and the 2 top quarks are decaying leptonically.

The analysis uses a data sample of proton-proton collisions at s√=13 TeV recorded with the ATLAS detector at the Large Hadron Collider in 2015, corresponding to a total integrated luminosity of 3.2 fb−1. The inclusive cross sections are extracted using likelihood fits to signal and control regions, resulting in σtt¯Z=0.9±0.3 pb and σtt¯W=1.5±0.8 pb, in agreement with the Standard Model predictions.

2. Analysis global information

3. Selection description

3.1 Trigger selection

Only events using single-electron OR single-muon triggers are accepted:

  • single-electron: one electron with pT > 24 GeV
  • single-muon: one muon with pT > 20 GeV

PS: This part cannot usually be described by the fast-simulation of detector.

3.2 Object definition

  • electron:
    • pT > 7 GeV,
    • |eta|<2.47. Are excluded electrons in the end-cap, i.e. 1.37 < |eta| < 1.52.
    • 'medium' likelihood identification.
    • transverse impact parameter: |d0|/err(d0) < 5.
    • longitudinal impact parameter: |z_0 x sin\theta | < 0.5 mm
    • isolation criterion:
      • sum PT of tracks in a cone around the electron (the electron track is subtracted) of size = min(10 GeV/PT,0.2) < 6% electron PT.
      • sum ET of clusters in a cone around the electron (the electron cluster is subtracted) of size DR=0.2 < 6% electron PT.

  • muon:
    • pT > 7 GeV
    • |eta|<2.4
    • 'medium' identification.
    • transverse impact parameter: |d0|/err(d0) < 3.
    • longitudinal impact parameter: |z_0 x sin\theta | < 0.5 mm
    • isolation criterion: sum PT of tracks in a cone around the muon (the muon track is subtracted) of size = min(10 GeV/PT,0.3) < 6% muon PT.

  • jet:
    • clustering: anti-kT with =0.4
    • pT > 25 GeV
    • |eta|<2.5
    • Pile-up rejection: jets with pT < 60 GeV and |η| < 2.4 are required to satisfy pile-up rejection criteria (JVT)
  • b-jet: MV2c20 algorithm. In simulation, the tagging algorithm gives a rejection factor of about 130 against light-quark and gluon jets, and about 4.5 against jets containing charm quarks.

  • preventing double-counting:
    • the closest jet within DR=0.2 cone of an electron is removed.
    • if, after the previous selection cut, the nearest jet of an electron is in DR=0.4 cone, then the electron is removed.
    • muons in a DR=0.4 cone around a given jet are removed. BUT if the jet has fewer than three associated tracks, the muon is kept.

3.3 Cut-flow chart

Common:

  • 0. Trigger confirmation: at least one lepton (muon/electron) with PT > 25 GeV

1. Analysis 1: SS-muons

  • 1.1. 2 muons of same charge, with PT > 25 GeV
  • 1.2. Events with additional leptons are vetoed
  • 1.3. MET > 40 GeV
  • 1.4. HT > 240 GeV
  • 1.5. At least 2 b-jets

2. Analysis 2: trilepton analysis

  • 2.1. Exactly 3 leptons
  • 2.2. 2 opposite sign, same flavour leptons, and
  • 2.2. Leading lepton with PT > 25 GeV, the other leptons with PT > 20 GeV
  • 2.3. Defining 4 signal regions according the final states:
    • 3l-Z-1b4j
    • 3l-Z-2b3j
    • 3l-Z-2b4j
    • 3l-noZ-2b
  • 2.4. Compatibility with a Z:
    • [3l-Z-1b4j]+[3l-Z-2b3j]+[3l-Z-2b4j]: compatible with a Z i.e. (invM-MZ)<10 GeV
    • [3l-noZ-2b]: not compatible with a Z i.e. (invM-MZ)>10 GeV
  • 2.5. Number of light jets (no b-jet):
    • [3l-Z-2b3j]: #jets = 3
    • [3l-Z-1b4j]+[3l-Z-3b4j]: #jets >= 4
    • [3l-noZ-2b]: #jets = 2, 3 or 4
  • 2.6. Number of b-jets:
    • [3l-Z-1b4j]: #b-jets = 1
    • [3l-Z-2b3j]+[3l-Z-3b4j]+[3l-noZ-2b]: #b-jets >= 2

3. Analysis 3: tetralepton analysis

  • 3.1. Exactly 4 leptons
  • 3.2. 2 pairs of opposite sign
  • 3.3. At least one pair must have same flavour leptons
  • 3.4. We define Z1 = reconstructed Z from opposite-sign same-flavour pair the closest to the Z mass. We define Z2 = reconstructed Z from the other pair (same-flavour or not).
  • 3.5. Defining 4 signal regions according the final states:
    • 4l-DF-1b
    • 4l-DF-2b
    • 4l-SF-1b
    • 4l-SF-2b
  • 3.5. Cut on PT34 = scalar sum of the transverse momenta of the leptons used in Z2:
    • [4l-DF-1b]: PT34 > 35 GeV
    • [4l-SF-1b]: PT34 > 25 GeV
  • 3.6. MET:
    • [4l-SF-1b]: If |MZ2-MZ| > 10 GeV then imposing MET > 40 GeV else imposing MET > 80 GeV
    • [4l-SF-2b]: If |MZ2-MZ| < 10 GeV then imposing MET > 40 GeV
  • 3.7. Number of b-jets:
    • [4l-DF-1b]+[4l-SF-1b]: #b-jets = 1
    • [4l-DF-2b]+[4l-SF-2b]: #b-jets >= 2

3.4 Definition of exotic observables

No exotic observables.

3. Detector simulation

Official Delphes package could be enough.

To improve the realism of the simulation, Delphes_Track could be used for cutting on D0. Is it necessary? To try...

4. Validation material

4.1 Monte-Carlo samples

  • LO (but xsections are normalized to NLO)
  • Model used: SM
    • Mtop = 172.5 GeV
    • Mhiggs = 125 GeV
    • processes = ttZ (including gamma*) & ttW (but no ttWW)
  • MC generator program (name + version): MG5_aMC
  • Shower program (name + version): Pythia v8.210
  • Tune of Pythia: A14 tune (see ​http://cds.cern.ch/record/1966419)
  • PDF set: NNPDF2.3LO
  • ME/PS merging????

4.2 Reference plots / cut-flows

  • Final number of signal events after the selection in each signal region:

​https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/TOPQ-2015-22/tabaux_06.pdf

  • Plots for Analysis 1: SS-muons
  • Plots for Analysis 2: trileptons
  • Plots for Analysis 3: tetraleptons
  • Rivet analyses

4. Validation results

5. Reinterpretation (optional)

6. Questions / issues

Note: See TracWiki for help on using the wiki.