ATLAS Workshop on Refinement of Local Hadronic Calibration and its Application to Top Physics

SESSION 4: tt(bar) selection, top mass reconstruction, QCD background, analysis tools

Conveners : D. Pallin , F. Spano'

General links :

RingbergWorkshop08

Agenda

ATLAS Calorimetry Calibration Workshops

Top Working Group

Top Meetings

Outline

Topics Summary

Top Mass measurement summary
Cross section summary

Ttbar Semileptonic channel

  1. Selection :
    1. Signal and BKG cross sections
    2. Triggers aspects ( D. Berge talk 10mn)
      1. Efficiency measurement, monitoring, redundancy
    3. Selection algorithms
      1. Basic Selection for xsec measurement
        1. X-sec Commissioning cuts
        2. b-tag impact
      2. Basic selection for mass measurement
        1. Mass Selection cuts; 2btag selection for mass measurement ; events after selection
        2. with loose b tag perf (2 b-tag+1-tag)
        3. No b-tag (at LHC start)
      3. Differences between cross section and mass selection
    4. Impact of Jet scale and pile-up on Jet Pt cut
      1. * Jet en scale impact for likelihood fit
    5. Impact of ETMiss
      1. ETMiss cut and QCD BKG for Xsec
        1. Cutting on ET miss vs electron fake rate
        2. Isolated lepton rate in multi jet events: latest and previous and critical points
          1. Communication on Isolated lepton rate in multi-jet events
          2. electron fake rate estimate and triangular cuts
          3. ETmiss calib
    6. Open issues
        1. Bkg studies
          1. QCD multijets 2-5 light jets ; bb+0,1,2,3 light jets and lepton fake rate
          2. W+bb w+cc (negligeable contribution)
        2. Etmiss calibration and use at LHC start
        3. use of different jet algorithm (cone,Kt) and different constituents (tower, Topocluster) +Loc Had calib( YES/NOT)
        4. Use locally hadronic calibrated clusters (and jets) in cross section
        5. Pile-up effect
  2. Reconstruction
    1. Reconstruction for cross section
    2. Association-Jet pairing: How to associate Jets to form the Hadronic Top
      1. Standard ; additional slide Methw1 ; Meth 1&2
      2. MVA
      3. Recons of the Leptonic Top
    3. Purification cuts/BKG substraction
      1. purification cuts for mass and for xsec
      2. MVA: Multivariate analysis
    4. Selection & Reconstruction of high pT Top (Elin Bergeas)
    5. Open issues
      1. Selection of a pure W sample to perform in-situ calibration (no b-tag)
      2. Kt/ cone ; Tower /Topcluster ; Local had calib( YES/NO) use for jets : differences for reconstruction
      3. comb BKG estimate with Evt mixing
  3. Extraction of measurements (cross section/mass)
    1. Cross section - T6 link
      1. three methods (jet multiplicity, cut and count, likelihood fit to mass estimator)
      2. Cross section systematics and Jet/EtMiss scale
    2. Mass - T9 link; material
      1. with optimal b tag performances
      2. with loose b tag perf
      3. at LHC start
      4. Mass systematics

Top Quarks at 10 -14 TeV

  1. Expectations on Cross sections and Kinematics material 1 material 2

Strategy for First Top measurements and feed back to the local Had calibration

  1. Two goals
    1. Improve jet/Et miss energy scale ( see In-Situ Session)
      1. jet scale
        1. material for light jet calibration material1 ; material2, material3
        2. JES and jet angular resolution?
        3. b-jet calibration from light jet calibration material
        4. JES and Mtop in one go
        5. jet energy resolution effect , pile-up effect
      2. ETmiss calibration
      3. Is there any step of the analysis that can validate local hadron calibration as a whole or in part?
    2. Get a robust measurement
      1. Top mass with Local Hadronic Calibration (T Barillari)
      2. What are the weakest points of the analyses? At what level are measurements more sensitive to Hadron calibration. How do we minimize the sensitivity to get the best initial measurement?
      3. What aspect of the local hadron calibration has the largest impact on the measurement? Classification, weighting, out of cluster, dead material, out of jet/, other physics corrections? What is critical, fragile ?
  2. Strategy at LHC start
  • make the measurement robust against calibration problems
  • and go towards a calibration scheme that will improve the measurement
  • use top to refine the day0 calibration
    1. 1st pass :
      1. Simple selection with/without btag
      2. W selection and evaluation of light jet scale
      3. b-jet scale obtained from light jet scale+ MC modelling the difference Light-b jets
    2. 2nd pass :
      1. Selection redone using in-situ calibration from pass 1
      2. Check important variables, #jets(light and b-jets)) and leptons, E,phi,eta, DRjj DRjjb , ETmiss,..
      3. Check calib
      4. Check with other selections procedures
      5. Measurements ?
        1. Determine X-sec from various methods (croos check)
        2. Top mass from Mjjb ; Cross-check b jet scale vs Mtop PDG
        3. check BKG with Evt mixing
        4. W mass :use different jet algo & cluster (tower, Topo) +-Loc Had Cal to give inputs to Loc Had Calib

Tools

  1. Top as a tool for performances studies
    1. Light jet scale
    2. B jet scale (not for Top mass measurement!)
    3. Trigger (use redundancy)
    4. Etmiss
  2. Analysis tools and format
    1. Compare fast and full simu: can we do had calib with fast simu?
    2. Is the infrastructure ready for putting things back from in-situ?
    3. What format for the initial analysis?
-- DominiquePallin - 03 Jun 2008 -- FrancescoSpano - 03 Jun 2008 -- FrancescoSpano - 04 Jun 2008 -- DominiquePallin - 05 Jun 2008 -- FrancescoSpano - 06 Jun 2008 -- DominiquePallin - 08 Jun 2008 -- FrancescoSpano - 10 Jun 2008 -- DominiquePallin - 12 Jun 2008 -- FrancescoSpano - 13 Jun 2008

Topic attachments
I Attachment History Action Size Date Who Comment
PowerPointppt 2008.06.13.TopTrigger.Ringberg.ppt r1 manage 1508.0 K 2008-06-13 - 11:28 DavidBerge  
PowerPointppt 2btag_selec.ppt r1 manage 21.5 K 2008-06-03 - 11:21 DominiquePallin 2b_mass-sel
PDFpdf BtagPerfXsec.pdf r2 r1 manage 205.4 K 2008-06-13 - 01:25 FrancescoSpano  
PDFpdf DioxadisKaylCriticalRingberg08.pdf r1 manage 52.4 K 2008-06-12 - 17:39 FrancescoSpano  
PDFpdf ETMissSelection.pdf r1 manage 219.7 K 2008-06-13 - 01:54 FrancescoSpano  
PowerPointppt EventMixing_cogneras_corr.ppt r1 manage 508.5 K 2008-06-05 - 09:16 DominiquePallin evtmix
PDFpdf ExtractMeasSystSpano.pdf r1 manage 356.1 K 2008-06-13 - 03:03 FrancescoSpano  
PDFpdf ExtractMeasusSpano.pdf r3 r2 r1 manage 110.2 K 2008-06-13 - 02:08 FrancescoSpano Ways to extract cross section
PowerPointppt FakeStudies_120308-1.ppt r1 manage 271.5 K 2008-06-12 - 21:52 DominiquePallin  
PowerPointppt JES_VVulpen.ppt r1 manage 209.0 K 2008-06-13 - 02:19 FrancescoSpano Jet e scale effects for likelihood fit
Unknown file formatpptx JESdppaf.pptx r1 manage 1159.8 K 2008-06-12 - 11:21 DominiquePallin  
PDFpdf JESpotpourridominique.pdf r1 manage 409.0 K 2008-06-12 - 16:23 DominiquePallin  
PowerPointppt JS_W_in_top-1.ppt r1 manage 728.0 K 2008-06-05 - 10:16 DominiquePallin  
PowerPointppt Plenary_top_121007-3.ppt r1 manage 249.5 K 2008-06-05 - 10:12 DominiquePallin  
PowerPointppt PresentationTop1.ppt r2 r1 manage 1203.5 K 2008-06-13 - 01:43 DominiquePallin  
PDFpdf ReconstructionXsecSpano.pdf r1 manage 40.9 K 2008-06-11 - 01:30 FrancescoSpano Simple reco for counting and lkl method
PDFpdf SpanoRingbergMain.pdf r2 r1 manage 20.9 K 2008-06-11 - 21:19 FrancescoSpano Basic summary of main points on xsec
PDFpdf Teresa-TopMass.pdf r1 manage 330.1 K 2008-06-12 - 23:58 DominiquePallin  
PDFpdf XSecSelectionHighPurSpano.pdf r3 r2 r1 manage 131.4 K 2008-06-13 - 00:46 FrancescoSpano Reconstruction and higher purity cuts
PDFpdf XSecSelectionSpano.pdf r4 r3 r2 r1 manage 302.3 K 2008-06-13 - 01:10 FrancescoSpano Basic xsec selection
PDFpdf ceanewmassmeas.pdf r1 manage 1159.3 K 2008-06-05 - 10:57 DominiquePallin  
Microsoft Word filedoc cross_sec.doc r1 manage 66.5 K 2008-06-03 - 09:50 DominiquePallin signal and BKG cross section
PowerPointppt crosssection_aie.ppt r1 manage 61.0 K 2008-06-03 - 11:20 DominiquePallin crosssecaie
PDFpdf ghodbane-15-05-2008dp.pdf r1 manage 2814.5 K 2008-06-08 - 10:12 DominiquePallin  
PowerPointppt introslide.ppt r2 r1 manage 32.5 K 2008-06-13 - 08:31 DominiquePallin  
PDFpdf jescscT2.pdf r1 manage 93.0 K 2008-06-05 - 10:33 DominiquePallin  
PDFpdf jesolag.pdf r1 manage 528.6 K 2008-06-05 - 10:09 DominiquePallin  
PowerPointppt massresonstruction.ppt r3 r2 r1 manage 751.0 K 2008-06-13 - 01:32 DominiquePallin  
PowerPointppt massselect.ppt r2 r1 manage 141.5 K 2008-06-13 - 01:34 DominiquePallin  
PowerPointppt metjesnikhef.ppt r1 manage 225.5 K 2008-06-08 - 09:07 DominiquePallin  
PowerPointppt mvaslide.ppt r1 manage 79.5 K 2008-06-06 - 11:34 DominiquePallin  
PDFpdf pileupmiller.pdf r1 manage 919.3 K 2008-06-08 - 09:07 DominiquePallin  
PDFpdf purif_cut2btag.pdf r1 manage 308.0 K 2008-06-03 - 15:28 DominiquePallin  
PDFpdf qcdkayl.pdf r1 manage 291.0 K 2008-06-05 - 10:09 DominiquePallin  
PowerPointppt reconslept_top.ppt r1 manage 73.5 K 2008-06-03 - 14:56 DominiquePallin  
PowerPointppt reconstophadmeth1and2.ppt r1 manage 276.5 K 2008-06-03 - 14:53 DominiquePallin  
PowerPointppt saclayrecons2btag.ppt r1 manage 76.5 K 2008-06-03 - 14:41 DominiquePallin  
Edit | Attach | Watch | Print version | History: r44 < r43 < r42 < r41 < r40 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r44 - 2008-06-13 - DavidBerge
 
    • 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 & 2008-2023 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