1 / 32

Mass Constraint on a SUSY Signal from Global Charge Asymmetry

This presentation discusses the use of global charge asymmetry to constrain the mass of a supersymmetric (SUSY) signal. Theoretical predictions and experimental measurements are presented, along with a breakdown of theoretical uncertainties. The goal is to establish mass templates with both theoretical and experimental uncertainties. The results provide valuable insights into the SUSY signal and its mass constraints.

lklug
Download Presentation

Mass Constraint on a SUSY Signal from Global Charge Asymmetry

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. S. Muanza, T. Serre CPPM Marseille, CNRS-IN2P3, France Mass Constraint on a SUSY Signalfrom Global Charge Asymmetry Terascale GDR Meeting @ Annecy

  2. Outline • Introduction • Reminders of the W test bench • Chosen SUSY Signal • Motivations to use AC • TheoreticalPrediction • Goal: Mass TemplateswithTheoreticalUncertainties • Mean: Parton Level Cross Sections • ExperimentalMeasurement • Goal: Mass TemplateswithExperimentalUnceratinties • Mean: ParticleLevel Simulation • Indirect Mass Constraint • Conclusions

  3. Introduction - Reminders on W - • In myprevious talk on thistopic, I presented a « New Method » to measure the W boson mass at ~10% accuracy • Reminders: • Global Charge Asymmetry: • Indirect W Mass Determination: • It worthwhileonly if: • the kinematic reconstruction has verypoorresolution • toomany final state particles are invisible

  4. Introduction - SUSY Signal (1) - • My main focus is to apply it to the following SUSY process in the trilepton+mET topology: • GCA could provide an additional handle:

  5. Introduction - SUSY Signal (2) - • SUSY Signal Scans: using Simplified Models

  6. 2. Theoretical Prediction - Parton Level Cross Sections - • Goal: establish a look-up table to translate an input GCA into a mass accounting for the theoreticaluncertainties • Wejust use LO cross sections calculatedwithProspino v2.1 with the following setup: • Phase Space: calls to VEGAS sufficient to getdsstat< 1 % • QCD Scales: • Parton LVL Cuts: none • Process ME: • Collider Energy: • Int. Luminosity: • PDFs: Had to use an hacked version by B. O’Leary to interface MRST and MSTW PDFs!

  7. 2. Theoretical Prediction - Theoretical Uncertainties - • Statistical Uncertainty: • Scale Uncertainty: • PDF Uncertainty:

  8. 2. Theoretical Prediction - Breakdown of Theoretical Uncertainties - • Example of SUSY Signal template data: CTEQ6L1

  9. 2. Theoretical Prediction - GCA Mass Templates - CTEQ6L1

  10. 2. Theoretical Prediction - GCA Enveloppes Fits - • Interpolate the 1s uncertainty curves • Same • Fit 2 curves with: Max Fit range: [15,1500] GeV, for all fits Central Min Tools 012, Stockholm

  11. 3. Experimental Measurement - Goals - • Extract the GCA of a given signal in a « realistic » environment • Study the experimental biases affecting AC: • Event selection (Trigger & Offline) • Background contamination • Account for experimental systematic uncertainties

  12. 3. Experimental Measurement - MC Production Tools - Herwig++ v2.5.2 HepMC v2.06.08 DELPHES v2.0 ROOT v5.28 Pythia v8.157 HPSS (RFIO) LHAPDF v5.7.1 Alpgen v2.14 • Event generation: • Generator: • Herwig++ v2.5.2 • For all SM processes, except… • Pythia v8.157 • For SUSY Signals • For Single Top • Alpgen v2.14 • VVV, W+c, W+cc, W+bb, W+tt; (Z+cc, Z+bb, Z+ttfailed)

  13. 3. Experimental Measurement - MC Production Setup - • PDF: MRST2007lomod • H++ UE Tune: • Name: « LHC-UE7-2 » • Ref: ATLAS_2010_S8894728 • Py8 UE Tune: • Name: « 2M » • Ref: arXiv:1003.2384 [hep-ph] • ME: LO • Normalization: LO xsect • No K-factors: QCD (LF), cc, bb, g+jets, gg (very low e) • NLO K-Factors: tt, t+X,VV, g+V (V=W or g*/Z) • NNLO K-Factors: V • Statistics: 108.1 M evts • Full details about: • MC Production: in appendix 1 • Fast Detector Simulation: in appendix 2

  14. 3. Experimental Measurement - Trigger & Event Skimming - Trigger Emulation Event Skimming • Isolated Leptons: • 1e (pT > 25 GeV) • 1m(pT > 20 GeV) • 2e(pT > « 2x15 » GeV) • 2m(pT > « 2x10 » GeV) • Isolated Photons: • 1g (pT > 60 GeV) • 2g(pT > « 2x20 » GeV) • Tau plus missing ET: • t+mET(pT > 35 & mET > 45 GeV) • Jet plus missing ET: • Jet+mET(pT > 70 & mET > 70 GeV) • Inclusive Jets: • 1j (pT > 400 GeV) • 3j (pT > « 3x165 » GeV) • 4j (pT > « 4x110 » GeV) • Goal: • Reduce specific data subsets to manageable size (that can be staged) • Event filtering: based on trigger bits • « 3lep »: • single lepton OR di-lepton

  15. 3. Experimental Measurement - Event Selection (1) - Preselection (1) Objects: • Muons • pT(m0) > 10 GeV • |h(m0)| < 2.40 • Muon Isolation: • Tracker: IsolFlag = 1 • Calo: ETRatio < 0.25 • Electrons • pT(e) > 10 GeV • |h(e)| < 1.37 or 1.53 < |h(e)| < 2.47 • Electron Isolation: • Tracker: IsolFlag = 1 • Calo: ETRatio < 1.5 No add’l tracks w/ pT > 2 GeV in a cone of DR=0.5 around m0

  16. 3. Experimental Measurement - Event Selection (2) - Preselection (2) Trilepton: • 3e • pT > 20, 20, 10 GeV • 3m • pT > 15, 15, 10 GeV • 1e+2m • Reject if: • pT(e0)<20 & pT(m1) < 15 GeV • 2e+1m • Reject if: • pT(e1)<20 & pT(m0) < 15 GeV Missing ET: • mET > 35 GeV Final Selection M4T: • m4T < 120 GeV or • m4T > 220 GeV

  17. 3. Experimental Measurement - Event Selection (3) -

  18. 3. Experimental Measurement - Experimental Mass Templates -

  19. 3. Experimental Measurement - Experimental Uncertainties - • Experimental systematic uncertainties: • Not derived by any DELPHES study • Take more realistic values from real data analyses • Very conservative examples (L < 35 pb-1): • Asymmetries: • in muon channel: 1.0% (ref: arXiv:1103.2929v1 [hep-ex]) • in electron channel: 1.5% (ref: ATLAS-CONF-2010-051) • S/B ratio: • in muon channel: 2.32% (ref: ATLAS-CONF-2011-041 ) • in electron channel: 2.80% (ref: ATLAS-CONF-2011-041 ) • We used 1.0% for GCA and 1.2% for S/B

  20. 3. Experimental Measurement - Extracting Signal GCA - • Now we need to extract the signal GCA: • without bias from the remaining background Ideal case More realistic case • Note: only ratios appear in extracted signal GCA (low systematics)

  21. 3. Experimental Measurement - Extracting Signal GCA Cont’d - • Now we need to account for all signal and background statistical and systematic uncertainties • We propagate those numerically into the previous formula. This enables to account simultaneously for all the uncertainties and their correlations.

  22. 3. Experimental Measurement - Extracting Signal GCA Cont’d - 10 M trials

  23. 4. Indirect Mass Constraint (1)

  24. 4. Indirect Mass Constraint (2)

  25. 4. Indirect Mass Constraint (3)

  26. 5. Conclusions • We introduced a new method exploiting the GCA at the LHC to contrain • We managed to reconstruct W mass indirectly: • Acceptable central values • Very bad resolution (as expected) • We managed to reconstruct the mass of a SUSY charged final state: • Acceptable central values • Even if the accuracy is not great, additional information is acquired • To be applied on real data, this method requires a very good MC modeling of the data

  27. BACK-UP

  28. IV. Indirect Mass Constraint - PDF RW in DELPHES - • The experimental mass templates are based on the MRST2007lomod PDF • To correctly interpret our results for any other PDF, either: • we redo the full MC production using that PDF or • we apply a PDF reweighting to the MRST samples: • Oops! I forgot to check wether the PDF variables (x1,x2,flav1,flav2,Q2) were stored in the DELPHES MC truth informations • Good news #1: they are stored by default in HepMC • Bad news: DELPHES ignore these infos! • … • Good news #2: I fixed this problem by: • retrieving the HepMC::PdfInfo object from the HepMC input file • stored the PDF variables into the « GEN » tree of DELPHES • you can find my fix at https://server06.fynu.ucl.ac.be/projects/delphes/ticket/44

  29. Appendix 2 -Fast Detector Simulation- • Detector datacard: ATLAS Objects Reconstruction Sub-Detectors Acceptance • Inner Tracker: • Central Calorimeter: • End-Cap Calorimeter: • Forward Calorimeter: • Muon Spectrometer: • Calorimeter Granularity: • Hadrons deposit in EM and HAD, separately smeared with EM and HAD resolutions • No Shower Simulation or Library! • Missing ET:

  30. Appendix 2 -Fast Detector Simulation- Objects Reconstruction Objects Reconstruction Thresholds • Jet Finder: FastJet v? • Algo: « Anti-kT » • Cone radius: DR=0.4 • Seed: pT > 1 GeV • Tracks: • Electrons: • Photons: • Jets: • Muons: • Taus: 3.51m 1.15m • Solenoidal Magnetic Field: • B=2 T • Applied on all stable charged particles • No simulation of toroidal magnetic field • Overall tracking efficiency:

  31. Appendix 2 -Fast Detector Simulation- • Objects Resolution: • Calorimeters:

  32. Appendix 2 -Fast Detector Simulation- DELPHES v2.0 • Output informations are stored into 3 different ROOT trees • « GEN »: MC truth (particles, ID, status, 4-p, charge, parenthood links,…) • « Analysis »: Reconstructed Objects (type, number, 4-p,…) • « Trigger »: • Pass or Fail for a list of trigger conditions • Based on a TCloneArray of the Analysis tree • No real separate trigger simulation: • Same calorimeter granularity • Same objects resolutions

More Related