Combined measurement of the Higgs boson mass in pp collisions at √ s=7 TeV and 8 TeV with the ATLAS and CMS experiments Hongtao Yang University of Wisconsin US LHC User Organization Annual Meeting Fermilab, Nov. 13, 2015
Motivation • Higgs boson mass (m H ) is a fundamental parameter of SM without a-priori expected value ‣ This measurement provides current best knowledge of m H based on LHC Run 1 data collected by ATLAS and CMS ➡ Input for future Higgs-related measurements/studies • Benefits of cooperation : valuable experience combining Higgs measurements between ATLAS and CMS A world record for the experiments at LHC length of the author list! ‣ Better understanding of the other experiment: performance, analysis strategies, systematic uncertainties etc. ‣ Accumulated experience benefits ongoing/future combinations/analyses Hongtao Yang (Wisconsin) US LUA 11/13/15 2
Input ATLAS analyses 200 weights / GeV -1 Ldt = 4.5 fb s = 7 TeV ∫ ATLAS Events / 2.5 GeV 180 -1 Ldt = 20.3 fb s = 8 TeV ∫ ATLAS 35 Data Data s/b weighted sum 160 Combined fit: Signal (m = 124.5 GeV = 1.66) → → µ H ZZ* 4 l H Mass measurement categories ∑ Signal+background 140 30 ∫ Background ZZ* Background -1 s = 7 TeV: Ldt = 4.5 fb Signal 120 Background Z+jets, t t ∫ -1 s = 8 TeV: Ldt = 20.3 fb 25 Systematic uncertainty 100 m H =124.51 ± 0.52 (stat.) ± 0.06 (syst.) GeV H → γγ 80 20 60 15 40 20 m H =125.98 ± 0.42 (stat.) ± 0.28 (syst.) GeV 10 0 8 weights - fitted bkg 6 4 5 2 0 -2 -4 0 -6 ∑ 80 90 100 110 120 130 140 150 160 170 -8 110 120 130 140 150 160 m [GeV] m [GeV] PRD 90, 052004 (2014) γ γ 4 l • m HATLAS = 125.36 ± 0.41 GeV = 125.36 ± 0.37 (stat.) ± 0.18 (syst.) GeV Hongtao Yang (Wisconsin) US LUA 11/13/15 3
Input CMS analyses -1 -1 -1 -1 19.7 fb (8 TeV) + 5.1 fb (7 TeV) s = 7 TeV, L = 5.1 fb ; s = 8 TeV, L = 19.7 fb CMS 3 4 10 × Events / 3 GeV S/(S+B) weighted events / GeV CMS Data 35 S/(S+B) weighted sum 3.5 H → γ γ Data Z+X 3 30 S+B fits (weighted sum) * Z ,ZZ B component γ 2.5 1 ± σ 25 2 ± σ m =126 GeV 2 H EPJ C 74 (2014) 3076 PRD 89, 092007 (2014) 1.5 20 1 + 0.26 = 1.14 µ − 0.23 15 0.5 m = 124.70 0.34 GeV ± m H =124.70 ± 0.31 (stat.) ± 0.15 (syst.) GeV H 0 10 200 B component subtracted 100 5 0 0 -100 80 100 120 140 160 180 110 115 120 125 130 135 140 145 150 m H = 125.59 +0.43-0.41 (stat.) m (GeV) m (GeV) 4 l γ γ +0.16-0.18 (syst.) GeV CMS • m H = 125.02 +0.29-0.31 GeV = 125.02 +0.26-0.27 (stat.) +0.14-0.15 (syst) GeV Hongtao Yang (Wisconsin) US LUA 11/13/15 4
Some details… • Workflow: 1. Construct combined likelihood model from input analyses 2. Perform maximum likelihood fits to derive measurement results • Correlations (especially between experiments): ‣ Production cross section ⨉ branching fraction ❖ Same physics: correlated between two experiments ❖ Minimal model dependency : uncorrelated between γγ and 4l. Also uncorrelated between ggF+ttH and VBF+VH within γγ ‣ Systematics: very limited correlation ❖ Two independent experiments : different detectors, different methodologies for systematic uncertainty evaluation ❖ Only correlate common theory uncertainties and part of luminosity uncertainties: trivial impact on mass measurement Hongtao Yang (Wisconsin) US LUA 11/13/15 5
Mass measurement results ATLAS + CMS , γγ + 4l m H = 125.09 ± 0.24 GeV = 125.09 ± 0.21 (stat.) ± 0.11 (syst.) GeV ATLAS and CMS Syst. Total Stat. LHC Run 1 Total Stat. Syst. ATLAS H 126.02 0.51 ( 0.43 0.27) GeV → γ γ ± ± ± CMS H 124.70 0.34 ( 0.31 0.15) GeV → γ γ ± ± ± ATLAS H ZZ 4 124.51 0.52 ( 0.52 0.04) GeV l → → ± ± ± CMS H ZZ 4 l 125.59 0.45 ( 0.42 0.17) GeV → → ± ± ± ATLAS + CMS 125.07 0.29 ( 0.25 0.14) GeV γ γ ± ± ± ATLAS + CMS 4 l 125.15 0.40 ( 0.37 0.15) GeV ± ± ± ATLAS + CMS +4 l 125.09 0.24 ( 0.21 0.11) GeV γ γ ± ± ± 123 124 125 126 127 128 129 m [GeV] PRL 114 (2015) 191803 H Hongtao Yang (Wisconsin) US LUA 11/13/15 6
m H vs. signal yield contours 3 =125.09 GeV) ATLAS H → γ γ ATLAS and CMS ATLAS H ZZ 4 l → → LHC Run 1 CMS H → γ γ 2.5 CMS H ZZ 4 l → → All combined H m Best fit ( SM 2 68% CL σ / σ = 1.5 S 1 PRL 114 (2015) 191803 0.5 124 124.5 125 125.5 126 126.5 127 Minimal model-dependency m [GeV] H Hongtao Yang (Wisconsin) US LUA 11/13/15 7
Systematic uncertainties ATLAS and CMS Uncertainty in ATLAS Uncertainty in CMS Uncertainty in LHC combined result combined result combined result LHC Run 1 ATLAS ECAL non-linearity / CMS photon non-linearity Material in front of ECAL ECAL longitudinal response ECAL lateral shower shape Photon energy resolution ATLAS H vertex & conversion → γ γ reconstruction Z ee calibration → CMS electron energy scale & resolution Muon momentum scale & resolution ATLAS H background modeling → γ γ Integrated luminosity ATLAS CMS Combined Additional experimental systematic uncertainties Observed Observed Observed Theory uncertainties Expected Expected Expected 0 0.05 0.1 0 0.05 0.1 0 0.02 0.04 0.06 PRL 114 (2015) 191803 m [GeV] δ H Hongtao Yang (Wisconsin) US LUA 11/13/15 8
Systematic uncertainties (cont’d) Uncertainty ATLAS and CMS Uncertainty in ATLAS Uncertainty in CMS Uncertainty in LHC contribution combined result combined result combined result LHC Run 1 ATLAS ECAL non-linearity / CMS photon non-linearity Energy/momentum Material in front of ECAL ECAL longitudinal response scale and ECAL lateral shower shape resolution 0.11 GeV Photon energy resolution uncertainties on ATLAS H vertex & conversion → γ γ reconstruction photons, electrons Z ee calibration → and muons CMS electron energy scale & resolution Muon momentum scale & resolution ATLAS H background modeling → γ γ Other experimental 0.02 GeV Integrated luminosity uncertainties ATLAS CMS Combined Additional experimental systematic uncertainties Observed Observed Observed 0.01 GeV Theory uncertainties Theory uncertainties Expected Expected Expected 0 0.05 0.1 0 0.05 0.1 0 0.02 0.04 0.06 PRL 114 (2015) 191803 m [GeV] δ H • Syst. uncertainty dominated by E/p scale and resolution of γ , e and μ • Theory uncertainty almost have no impact on results ✴ Interference between signal and continuum bkg. not considered though Hongtao Yang (Wisconsin) US LUA 11/13/15 9
Conclusion ATLAS+CMS combined measurement of the Higgs • boson mass in the H → γγ and H → 4l channels based on LHC Run 1 data m H = 125.09 ± 0.24 GeV = 125.09 ± 0.21 (stat.) ± 0.11 (syst.) GeV ‣ Total uncertainty dominated by statistical uncertainty ‣ Systematic uncertainty dominated by energy/ momentum scale and resolution uncertainties • Good consistency among input analyses • Already achieved <2‰ precision. Stay tuned for even better results from Run 2 data! Hongtao Yang (Wisconsin) US LUA 11/13/15 10
Back up Hongtao Yang (Wisconsin) 11 US LUA 11/13/15
ATLAS photon reconstruction ATLAS-CONF-2011-161 CERN/LHCC/96-041 (1996) • Liquid Argon sampling calorimeter with 3 layers (+ pre-sampler for | η |<1.8): ‣ Provides longitudinal shower information • Photon conversion reconstructed by matching tracks consistent with conversion • Photon pointing used for photon vertex selection: Improves resolution at high pile-up Hongtao Yang (Wisconsin) US LUA 11/13/15 12
CMS photon reconstruction E 3 ⨉ 3 E SC CERN-LHCC-2006-001 • High granularity PbWO 4 crystal layout • Sliding window / dynamic clustering algorithm to reconstruct bremsstrahlung and converted photons (“superclusters” in CMS) ‣ Lateral shower shape distinguishes converted from unconverted photon • Rely on tracks (recoil) to assign photon vertex Hongtao Yang (Wisconsin) US LUA 11/13/15 13
ATLAS photon calibration 0.02 3000 Scale Events / GeV ∫ -1 Unconverted photons ATLAS s = 8 TeV, Ldt = 20.3 fb 0.015 2500 Data ∆ Unconverted photons 0.01 Calibration uncertainty Z , data → µ µ γ 2000 0.005 Z , MC → µ µ γ Z ee , data → γ 0 1500 Z ee , MC → γ -0.005 1000 PRD 90, 052004 (2014) -0.01 500 -0.015 ∫ -1 ATLAS s = 8 TeV, L d t = 20.3 fb -0.02 0 10 20 30 40 50 60 60 70 80 90 100 110 120 E [GeV] EPJ C 74 (2014) 3071 m [GeV] T ll γ • Monte Carlo based energy calibration using multivariate techniques • Additional corrections from Z → ee MC/data comparison • Validate with real photons in data Hongtao Yang (Wisconsin) US LUA 11/13/15 14
Recommend
More recommend
