Properties of the Higgs Boson

Transcription

1 Properties of the Higgs Boson Rishi Patel Rutgers University On behalf of the CMS and ATLAS Collaborations Wednesday, August 4, 3

2 Higgs at the LHC One of the goals of the LHC physics program is to unravel the origin of Electroweak symmetry breaking: discovery fingerprint Higgs? Breakthrough: Discovery by CMS and ATLAS of a new boson discovery fingerprint FUTURE: 4 TeV SNOWMASS Projections Higgs BR Total Uncert [%] "(pp HX) [pb] From LHC cross sections Group cc M H [GeV] µµ bb pp H (NNLONNLL QCD NLO EW) gg "" Z" pp qqh (NNLO QCD NLO EW) pp WH (NNLO QCD NLO EW) pp ZH (NNLO QCD NLO EW) pp tth (NLO QCD) WW ZZ LHC HIGGS XS WG 3 M H [GeV] s= 8 TeV 5 GeV : Many available production modes and decay channels to test SM compatibility With the full Run data (~3/fb) the experiments can test the compatibility of the new boson with the prevailing theory: Standard model Higgs boson LHC HIGGS XS WG Wednesday, August 4, 3

3 Standard Model Higgs free parameter: μ=mh SM Higgs: v=μ/ λ=mw /g Tree level COMPATIBILITY: Compare the SM prediction to the for different quantities: signal strength:σobs x BR/σSM x BRSM coupling scale:κ gsm=g Loop level Standard... H, something new? Presence of non-sm particles in the loop 3 Wednesday, August 4, 3

4 Mass Measurement / SM Combining channels For Mass Measurement Combined H # "" H # ZZ m X ± CMS Preliminary s = 7 TeV, L $ 5. fb s = 8 TeV, L $ 9.6 fb H # "" q = log CMS H # ZZ L(µ s(θ)b(θ)) L(b(θ)) (GeV) CMS: m H = 5.7 ±.3(sys) ±.3(stat) ATLAS: m H = 5.5 ±.(sys).5.6 (stat) Signal strength (µ) ATLAS ATLAS Preliminary s = 7 TeV: Ldt = fb s = 8 TeV: Ldt =.7 fb Combined H γ γ H ZZ 4l Best fit 68% CL 95% CL m H [GeV] CMS: m H = 5.7 ±.3(sys) ±.3(stat) ATLAS: m H = 5.5 ±.(sys).5.6 (stat) For current data: Δm~±.5GeV but projections at 3/ fb (3/fb) at 4TeV show Δm~MeV (5MeV) based on Snomass projections MODEL DEPENDENT MEASUREMENT All productions and BR are constrained to the SM predictions. - ln L CMS Preliminary s = 7 TeV, L " 5. fb s = 8 TeV, L " 9.6 fb H # $$ µ H # ZZ, µ (ggh,tth), $$ µ (VBF,VH) $$ ZZ CLs=95% with syst. no syst. CLs=68% m X (GeV) Figure : D-scans of the test statistic q(m X ) versus the boson mass m X of the γγ and 4l final states. Best The solid fit mass line is obtained compatible with allbetter nuisance and hence includes both statistical than and. systematic GeV with uncertainties. the model The das with all nuisance parameters fixed to their best-fit values and hence inc independent uncertainties. The crossings with the thick (thin) horizontal lines defin interval for the measured mass. Wednesday, August 4, Compatibility of the state with the SM Higgs bos

5 Combined Signal Strengths Simultaneously analyze all selected data across all decay modes and measure the overall deviation from the SM cross-section ATLAS: Combined µ=.3±. CMS: Combined µ=.8±.4 ATLAS Preliminary W,Z H bb H $$ s = 7 TeV: %Ldt = 4.6 fb s = 8 TeV: %Ldt = 3 fb H WW H "" s = 7 TeV: %Ldt = 4.8 fb s = 8 TeV: %Ldt =.7 fb H ZZ Combined l#l# 4l s = 7 TeV: %Ldt = fb s = 8 TeV: %Ldt = fb s = 7 TeV: %Ldt = 4.7 fb s = 8 TeV: %Ldt = 3 fb s = 7 TeV: %Ldt = 4.6 fb s = 8 TeV: %Ldt =.7 fb s = 7 TeV: %Ldt = 4.6 fb s = 8 TeV: %Ldt =.7 fb NOW: High sensitivity decay modes basically drive the combination (~5% precision on combined signal strength) AT 4TeV: At high luminosity 3/fb, less sensitive decay modes have much smaller uncertainties. The combined signal strength will be even more precise Wednesday, August 4, 3 µ =.3 ±. m H = 5.5 GeV Signal strength (µ) CMS Projection Combined µ =.8 ± H " bb H " $$ H " ## µ =.5 ±.6 µ =. ±.4 µ =.77 ±.7 H " WW µ =.68 ±. H " ZZ Expected uncertainties on Higgs boson signal strength H " H " WW H " ZZ H " bb H " # #.4 µ =.9 ±.8 s = 7 TeV, L % 5. fb CMS Preliminary p =.65 SM s = 8 TeV, L % 9.6 fb m H = 5.7 GeV Best fit / SM 3 fb at 3 fb at s = 4 TeV Scenario s = 4 TeV Scenario expected uncertainty 5

6 Production Modes Combined µ=.8±.4 Combined µ =.8 ±.4 s = 7 TeV, L " 5. fb CMS Preliminary p =.5 SM s = 8 TeV, L " 9.6 fb m H = 5.7 GeV µ VBF,VH 6 CMS Preliminary s = 7 TeV, L $ 5. fb s = 8 TeV, L $ 9.6 fb H " H " WW H " ZZ Untagged µ =.78 ±.6 4 H " bb H " ## VBF tagged µ =. ±.34 VH tagged µ =. ±.49 tth tagged µ = -.5 ±.86 D scan: Fermion coupling to Higgs vs. Vector Boson coupling: Each contour is for a - ln L(µ s(θ)b(θ)) different q = logbr/brsm so difficult to combine L(b(θ)) 8 CMS: m H = 5.7 ±.3(sys) ±.3(stat) 6 ATLAS: m H = 5.5 ±.(sys).5.6 (stat) D Scan of ratio: Branching ratio of each decay cancels ATLAS: µ ggftth µ V BF VH = Best fit / SM 4 4 ATLAS s = 7 TeV: s = 8 TeV: = 5.5 GeV m H combined SM expected Preliminary H # " " # 4l # l$l$ &Ldt = fb &Ldt = 3-.7 fb H # ZZ H # WW H # % % µ / µ VBFVH ggftth 3 µ ggh,tth Projection of 3/fb to 4TeV predicts a much tighter precision of ~% 6 Wednesday, August 4, 3 (a) (b)

7 σ BR(ii L(b(θ)) H ff) = ΓH ATLAS: µggf tth µv BF V H Coupling Scale Factors µ ggf tth.7 CMS: mh = 5.7 ±.3(sys) ±.3(stat) ATLAS: =. L(µ,θµ ) µv BF V H.5.5 qµ = ATLAS: mhlog = 5.5 L(µ,θ )±.(sys).6 (stat) ps σ BR(ii H f f ) = CLs = p σ BR(ii H f f ) = σii Γf f ΓH qµ = for log Loop level γγ: b µggf tth.7 L(µ,θµ ) ATLAS: µ =. q =.5 log L(µ,θ ) V BF V Hµ f f ps σ Γ CL = σ W,Z BR(ii W,Z H s f f ) p =b iiγ f f H L(µ,θµ ) L(µ,θ ) Tree g mlevel amplitudes: ps interference: CL = m g Γjj s (mjj κjjg)f mfκf SM κ Γ g κ m jj jj W,Z W,Z W,Z v (mjj κjj ) Γjj channel can be represented a vas product of coupling scale factors κjj ΓSM Γ ακ βκ γγ W t jj pb gf mf κf gw,z κw,z mw,z 8 coef. fixed to(msm values jj κjj ) SM Γjj κ jj Γjj v ΓH =.75κf.5κV Assume: SM tensor structure JP= and SM BR to fermions/vector bosons: s = 7 TeV, L % 5. fb s = 8 TeV, L % 9.6 fb CMS Preliminary f (common scale SM Best fit for all factor 68% CL fermions) 95% CL.L. 95% C H"## $$ H " H " bb - F ( ) V Wednesday, [, ] August 4, 3 F H " WW.3 κw=κz V Bkg. only H " ZZ (common scale factor for all vector bosons) κv = Fermiophobic κt=κb.... SM Higgs κf=. = V Figureat8:68% The 68% CL contours for individual ch ATLASκF [.73,.7] κv [.5,.] C.L. all combination (solid line) for the (κv, κf ) param (b) CMSκ F [.6,.33] κv [.74,.6] at 95% C.L. values. The thin contour shows the 95% CL rang 7 shows the SM point (κv, κf ) = (, ). The left plot (,) and (, ), the right plot shows the positive q 9 ATLAS Preliminary [ V, F]

8 SM Compatibility Tests Custodial symmetry: Issue is Γγγ depends on κw CMS and ATLAS: Decouple the event rate of H γγ from κw/κz by introducing additional free parameters in the likelihood 8 W/Z coupling to Higgs: gz/gw CMS Preliminary s = 7 TeV, L # 5. fb s = 8 TeV, L # 9.6 fb 5. ATLAS Preliminary WZ, $ Z, $ f Observed s = 7 TeV, %Ldt = fb Exp. for SM H 8 4. s = 8 TeV, %Ldt = 3-.7 fb Figure 8: The 68% CL contours 3. for individual channels (coloured swaths) and for the overall combination (solid line) for the (κ V, κ f ) parameters. The cross indicates. SM 5 ATLAS Preliminary the global best-fit.5 s = 7 TeV, #Ldt = fb values. The thin contour shows the 95% CL range for the combination. The 4 yellow diamond s = 8 TeV, #Ldt = 3-.7 fb shows the SM point (κ V, κ. f ) = (, ). The left plot shows the likelihood scan 3.8 in two quadrants (,) and (, ), the right plot.5 shows the positive quadrant only..6 - " ln L. ) - ln "( WZ.4 [ WZ, FZ, $Z,# ZZ ] Observed SM expected (a) WZ 8. CMS Preliminary.8 s = 7 TeV, L # 5. fb s = 8 TeV, L # 9.6 fb 7 ATLAS Preliminary " WZ, ". SM 6 g.6 Figure 6: Likelihood scan versus λ WZ, the ratio of the.8 s = 7 TeV, #Ldt = fb Best fit 5 couplings to W and Z bosons. (Left) 4.9 s = 8 TeV,.#Ldt. = fb.568%.6cl from untagged pp H WW and inclusive pp.6 H ZZ searches, and assuming SM.8 95% CL " couplings to fermions. (Right) from the combination of.4all channels, profiling the coupling to fermions. The solid curve is the data. The dashed line indicates the expected median results.6 (a). in the presence of the SM Higgs boson. Crossings with the horizontal thick and thin red lines.4 denote the 68% CL and 95% CL intervals.. gg H..8 Figure : Fits for benchmark models described in Equation (44) probing.6 particles in the H γγ and gg H loops, assuming no sizeable extra con (a) correlation.8 of the coupling scale factors κ.4 γ and κ g ; (b) coupling scal (c) coupling. 4.3 Compatibility of the state with the.6 scale factor κ g (κ γ is profiled). The dashed curves in (b) and (c) SM Higgs boson hypothesis 9 Probe Loop Corrections: H γγ Scenario New particles contribute negligibly to the total width: Γtotal=ΓSM Fit kg kγ Scenario Allow new particles to contribute to the total width: Γtotal=ΓSMΓBSM Fit kg, kγ, ΓBSM Wednesday, August 4, 3 " g " CMS Preliminary s = 7 TeV, L " 5. fb s = 8 TeV, L " 9.6 fb ATLAS Preliminary = 7 TeV, L " Figure 9: The D likelihood scan for the5.κ s 5. fb g and # $, # κ g, γ BR parameters, assuming that9γ BSM BSM Observed s = = 7 TeV,, #, i.e. g $Ldt, BR= BSM (a) fb no new Higgs boson decay modes are open. 4.5 The cross indicates the best-fit values. 8 The solid, Exp. for SM H s = 8 TeV,.5 $Ldt = 3-.7 fb dashed and dotted contours show the 68%, 4. 95% and 99.7% CL regions, respectively. 7 The yellow - ln L Figure 7: The D likelihood of the κ V and κ f parameters. The cross indicates the best-fit values. BR BSM ATLAS Preliminary 9 BR BSM [" #," g,b ] i,u The solid, dashed and dotted contours show the 68%, 95% and 99.7% CL regions, respectively. Figure : (Left) The likelihood scan versus BR The yellow diamond shows the SM point (κ, κ )=(, ). The left plot shows the likelihood BSM = Γ s = BSM /Γ 7 TeV, tot. The $Ldt solid = (a) curve is fbthe data and Observed 8 " g g 3. CMS Preliminary s = 8 TeV, L " 9.6 fb diamond shows the SM point (κ γ, κ g ) = 3.5 (, ). The partial widths associated with6 the tree-level. production processes and decay modes are assumed to be unaltered Figure (κ = ). : Fits for benchmark models described in Equation (44) 3. 5 particles in the H γγ and gg H loops, assuming no sizeable e (a) correlation of the coupling scale factors κ γ and κ g ; (b) coup " g ) 3 (c) coupling scale. factor κ g (κ γ is profiled). The dashed curves in (b ) i,u ln (B i,u - ln (B.5 Best fit 68% CL 95% CL [" #," g,b ] i,u Observed SM expected B i,u ) - ln #(" ) g - ln "( " ) - ln #(" ) g - ln "(

9 SUMMARY: Couplings and Total Width Ratios of couplings requires no assumption on the total width Can include total width including extra contributions: f Γ H = Γ SM Γ BSM V BR SM = Γ BSM Γ H For 3/fb at 4TeV the statistical uncertainty are below % Theory systematics most important: QCD scale, pdf uncertainties, BR uncertainties " V " b " # " t " g " BR BSM s = 7 TeV, L $ 5. fb CMS Preliminary s = 8 TeV, L $ 9.6 fb [ " V $ ] 68% CL 95% CL p SM p SM =.78 = parameter value Higgs boson coupling parameters are show ATLAS Preliminary s = 7 TeV, $Ldt = fb ± % ± % " " V F # " FV VV # # FV " V VV # # WZ " FZ ZZ ## WZ Z # " FZ " ZZ g " " g " -B i,u m H = 5.5 GeV s = 8 TeV, $Ldt = 3-.7 fb parameter value CMS Projection Expected uncertainties on Higgs boson couplings " " W " Z " g " b " t " # Wednesday, August 4, 3 3 fb at 3 fb at s = 4 TeV Scenario s = 4 TeV Scenario all sys expected uncertainty CMS Projection CMS Projection Expected uncertainties on Higgs boson Expected couplingsuncertainties on Higgs boson couplings " " W " Z " g " b " t " # " " W " Z " g " b " t " # 3 fb at 3 fb s = 4 TeV Scenario at s = 4 3 TeV fb Scenario at s = 4TeV Scenario 3 fb at s = 4 TeV No Theory Unc. no theory sys expected expected uncertainty uncertainty 9

10 Pseudoexperiments Spin Hypothesis Test Test Spin SM Higgs Hypothesis vs a Spin M hypothesis. CMS preliminary s = 7 TeV, L = 5. fb ZZ: CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb Fully reconstructed 4-lepton final state: s = 8 TeV, L = 9.6 fb - CMS data Figure 3: Distributions of D J P with a requirement D bkg >.5. Distributions in data (points with error bars) and expectations for background and signal are shown. Six alternative hypotheses are tested from top to bottom and left to right: J P =, h,,, m(gg), m(q q). Pseudoexperiments m(gg) CMS data CMS data.8.8. CMS data WW:Not a fully reconstructed final state but have. angles computed from the leptons MVA classifier is trained with final.4state observables... Pseudoexperiments ln(l - / L ) CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb Pseudoexperiments ln(l - / L ) - ln(l - / L -3 - ) 3 - ln(l / L ) h - CMS data CMS data Rules out - Confirms not h Pseudoexperiments Pseudoexperiments Pseudoexperiments ln(l / L ) CMS.6 CMS preliminary preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb s = TeV, 5. fb s = 8 TeV, L = 9.6 fb m(gg) CMS data ln(l / L ) -3 - m(gg) 3 - ln(l / L ) - (pure gg m) Pseudoexperiments Pseudoexperiments.6 CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb ln(l / L ) m(gg). CMS preliminary s = 7 TeV, L = 5. fb ln(l / L ) m(qq ) (pure qq m) s = 8 TeV, L = 9.6 fb m(qq) CMS data Figure 4: Distribution of q = ln(l J P/L SM ) for two signal types ( represented by the yellow histogram and alternative J P hypothesis by the blue histogram) for m H = 6 GeV shown with a large number of generated experiments. The arrow indicates the value. Six alternative hypotheses are tested from top to bottom and left to right: J P =, h,,, mgg, mq q. CMS WW/ZZ Combination: Exclude m for pure ggh model at.84σ Wednesday, August 4, 3 Spin M use graviton model simulation produced via gluon fusion and quark-antiquark (giving different polarization) ZZ predictions: 4.4 Tests of different spin-parity hypotheses Probability density CMS preliminary s = 7 TeV, L = 5. fb s = 8 TeV, L = 9.6 fb ln(l / L ) m(gg) m(gg) CMS data Figure 4: Post-fit model distributions of the test statistic comparing and m(gg) in the best fit to the data. The value is indi disfavours the m(gg) signal hypothesis with a CL s value of.6%.

11 Combined Spin Tests 3 γγ CS frame:.6 ATLAS Preliminary s = 8 TeV, L dt =.7 fb ATLAS: WW,.4ZZ, H# WW # e"µ"/µ"e" γγ jets γγ sensitive at low qq admixture WW sensitive at high qq admixture P J = (SM) hypothesis ATLAS Preliminary P J =, %qq hypothesis L dt =.7 fb H γ γ, s = 8 TeV P J = (SM) hypothesis ATLAS Preliminary P J =, 75%qq hypothesis L dt =.7 fb H γ γ, s = 8 TeV. use angle between the photons in the collins-soper frame: Spin ln(l()/l()) ln(l()/l()) the angular distribution is isotropic as opposed ln(l()/l()) to spin.8. P Figure 9: Expected ATLASdistributions Preliminary of ln(l()/l()) f = % the.6 qq logarithm ATLASofPreliminary the ratio of profiled f qq = 5% likelihoods Figure.6 9: Expected ATLAS distributions of ln(l()/l()) the logarithm of the ratio of profiled likelihoods J = (SM) hyp P P J = (SM) hypothesis ATLAS Preliminary J = (SM) hypothesis ATLAS Preliminary 3 P J =, 5%qq under the spin- and spin- hypotheses in the presence of spin- (blue distributions) P 6 J =, 5%qq hypothesis 3 P % or spin- under (red.6 the spin- 3 Preliminary f qq = 5% and spin- hypotheses L 6. dt =.7 fb in the 4 ATLAS Preliminary f.6. qq = 75% ATLAS presence ofjspin- =, 5%qq (blue hypothesis.65 distributions) orspin- (red L dt =.7 fb.4 4 Preliminary f 4. 4 s = 8 TeV, L dt =.7 fb H [5] s = 8 TeV,.4 5 H γ γ distributions).4 signals # for e"µ"/µ"e" L dt =.7 fb H [5] H# WW jets H [5] various admixtures of gg-q q production of the # e"µ"/µ"e" H# WW jets H [5].4 s = 8 TeV, L dt =.7 fb H qq = % [5] distributions) H# signals WW # e"µ"/µ"e" for various jets H [5].8 s = 8 TeV, L dt =.7 fb H [5].5 s = 8 TeV, admixtures of gg-q q production H# WW of # e"µ"/µ"e" the spin- jets H [5] L dt =.7 fb signal: % gg % q q H [5] H# WW # e"µ"/µ"e" jets H [5] spin- signal: % gg % q q. 5 H γ γ, s = 8 TeV 5 H γ γ, s = 8 TeV in (a), 5%..6 in (a), 5% gg 75% q q in (b), 5% gg 5% q q in (c), 75% gg 5% q q in (d). The values are. gg 75% q q in (b), 5% gg 5% q q in (c), 75% gg 5% q q in (d). The values are. 4.4 indicated by a vertical line. The coloured areas correspond. to the integrals of the expected distributions indicated. by a vertical line. The coloured areas correspond to the integrals of the expected distributions 5.. up 3 to the values and are used to compute the 5p-values for the rejection of each hypothesis..5 up to the values and are used to compute the p-values.8 for the rejection of each hypothesis ln(l()/l()) ln(l()/l()) ( -5 5 (c) (d) Figure 9: Expected distributions of ln(l()/l()) the logarithm of the log(l(h )/L(H 5 ) ) -5 5 log(l(h )/L(H ) ) log(l(h )/L(H 5 ) ) -5 5 log(l(h )/L(H ) ) log(l(h )/L(H 5 ) ) Figure 9: Expected distributions of ln(l()/l()) the logarithm of the ratio 4of profiled likelihoods under the spin- and spin- hypotheses in the presence of spin- (blu distributions) signals for various admixtures of gg-q q production of the sp under the spin- and spin- hypotheses in the presence of spin- (blue distributions) Figure : Test orstatistics spin- (red for (red) and m (blue) as well as the value (solid line) for the in (a), 5% gg 75% q q (b), 5% gg 5% q q in (c), 75% gg 5% q q i distributions) signals for various admixtures of gg-q q 8 production of the spin- signal: ATLAS five different % gg % Preliminary m production working points: f q q q q = %% (top left), f q q = 5% (top right),. ATLAS Preliminary f 8 3 indicated by a vertical line. The coloured areas correspond to the integral in 4 (a), 5% ATLAS gg 75% Preliminary qq = % f f qq = 5% ATLAS Preliminary f = 75% q q in (b), 5% gg 5% q q in (c), 75% gg 5% q q in (d). 4 q q = 5% The ATLAS (middle left), values Preliminary f q q = 75% (middle right), and f q q = % (bottom). The median of each of the. qq.5 s = 8 TeV, are up to the values and used to compute the p-values for the reje H [5] s = 8 TeV, indicated by a vertical line. The coloured areas correspond Spin to the integrals of the expected H ZZ* distributions 4l Spin H [5].8 L dt =.7 fb L dt =.7 fb H [5] distributions is indicated by a dashed line. The shaded areas correspond to the p H [5] -values. The H# WW # e"µ"/µ"e" jets H [5] H# WW # e"µ"/µ"e" jets H [5] H ZZ* 4l distributions H ZZ* are normalised 4l to unit area. up to the values and are used to compute the p-values for the rejection of eachs hypothesis. = 7 TeV: Ldt = 4.6 fb..6. s = 7 TeV: Ldt = 4.6 fb Signal hypothesis σ s = 7 s TeV: = 8 TeV: Ldt = Ldt 4.6 = fb.7 fb Signal hypothesis σ.4. s = 8 TeV: 3 Ldt =.7 fb.5 s = 8 TeV: σ 3 Ldt =.7 fb σ. P H γγ PCL expected s H γγ J =.8 H H s γγ J =. = 8 TeV: Ldt =.7 fb H P. (assuming J = ).6.8 s = 8 TeV: Ldt =.7 fb s = 8 TeV: P J = H WW* Ldt =.7 fb eνµν/µνeν P.6 H J = H WW* eνµν/µνeν H.4.5 H s WW* = 8 TeV: Ldt eνµν/µνeν =.7 fb.4 s = 8 TeV: Ldt =.7 fb s = 8 TeV: Ldt =.7 fb. σ σ log(l(h )/L(H 5 log(l(h ) ) -5 5 log(l(h )/L(H 5 )/L(H 5 ) ) ) ) - Figure : Test statistics distributions for (red) and m (blue) as well as the value (solid line) for the five different m production working points: f q q = %% (top left), f q q = 5% (top right), -3 3σ ATLAS Preliminary f qq = % f q q = 5% (middle left), f q q = 75% (middle right), and f q q = % (bottom). The median of each of the.5 s = 8 TeV, L dt =.7 fb distributions is indicated by a dashed line. The shaded areas correspond to the p -values. The H [5] H# WW # e"µ"/µ"e" jets H [5] distributions are normalised to unit area σ 5 4 P J = (SM) hypothesis ATLAS Preliminary P J =, %qq hypothesis L dt =.7 fb H γ γ, s = 8 TeV Arbitrary Normalisation ATLAS Preliminary.6 s = 8 TeV, L dt =.7 fb P H# WW # e"µ"/µ"e" jets J = (SM) hypothesis.4 ATLAS Preliminary P J =, 75%qq hypothesis. L dt =.7 fb H γ γ, s = 8 TeV Table : Expected and p values for the J P = -5 and J P ln(l()/l()) = 5 log(l(h hypotheses )/L(H 5 ) ) -5 5 log(l(h as a function of the )/L(H 5 ) ) (a) (b).8 fraction f (a) (b) q q of the q q ATLAS spin- Preliminary production f = %.6 mechanism. ATLAS qq The Preliminary values fare qq = 5% calculated for the combination of the P P J = (SM) hypothesis ATLAS Preliminary J = (SM) hypothesis H ATLASγγ, Preliminary H WW.6 s = 8 TeV, L dt =.7 fb lνlν and ZZ H [5] s = 8 TeV,.4.4 4l final states. The CL P P P P 5 J = J =(SM) (SM) hypothesis ATLAS Preliminary ATLAS Preliminary P J =J = (SM) (SM) hypothesis ATLAS Preliminary P 6 J =, 5%qq hypothesis 3 P s values defined in the text are # e"µ"/µ"e" L dt =.7 fb. H [5].6 ATLAS H# WW jets H [5] # e"µ"/µ"e" Preliminary f H# WW jets H [5] qq = 5%. ATLAS Preliminary f qq = 75% s = 8 TeV,..4 L dt =.7 fb H [5] s = 8 TeV, 5 J = (SM) hypothesis ATLAS Preliminary P P J =, %qq P 6 J =, 5%qq hypothesis 3 P P J =, %qq hypothesis L dt =.7 fb J =, 75%qq L dt =.7 fb J =, 5%qq hypothesis H# WW # e"µ"/µ"e" jets H [5].8 L dt =.7 fb H [5] H# WW # e"µ"/µ"e" jets H [5] L dt =.7 fb J =, 5%qq hypothesis L dt =.7 fb. L dt =.7 fb L dt =.7 fb also presented. L dt =.7 fb 5, s = 8 TeV H γ γ 5 H γ γ, s = 8 TeV H H γ γ γ γ,, s s = 8 = TeV 8 TeV 8 5H H γ γ γ γ,, s = s 8 TeV = 8 TeV 4 H γ γ.4, s = 8 TeV Spin- assumed Spin- assumed f q q 3 5 obs. p 4 (J P.8.6 = ) obs. p (J P = ) CL s (J P = ) exp. p (J P = ) exp. p (J P = ) 5 % % ln(l()/l()) ln(l()/l()) ln(l()/l()) ln(l()/l()) ln(l()/l()) ln(l()/l()) ln(l()/l()) 5%.5 log(l(h )/L(H 5 (c) (d) (a) 3.5 ) 3 ) -5 5 log(l(h )/L(H 5-5 (c) (b).85 ) ) 5 log(l(h 9. )/L(H ) ) -5 5 log(l(h )/L(H 5 Increasing qq admixture ) ) (d) Expected exclusion of spin m depends on fqq very weakly is consistent with and m is excluded at 99.9 % confidence level Wednesday, August 4, log(l(h )/L(H 5 ) ) 5% )) )/L(H log(l(h (a) f qq [%] ) )/L(H P )) (J = log(l(h CL s f qq [%] [%] (b) f qq = % H [5] H [5].6 ATLAS Preliminary s = 8 TeV, L dt =.7 fb H# WW # e"µ"/µ"e" jets f qq f qq = 5% H [5] H [5]

12 New Studies Use high signal yield mode: H γγ to probe kinematic properties of production/decay: Extract a signal yield for bins of a kinematic variable correct yield for acceptance x efficiency, resolution etc. to compare to theory predictions (Left) Compare data to simulation (NLO and NNLO for ggh) Chi: NLO=.55 and NNLO=.39. (Right) signal strength in bins of CS angle offers a potentially model independent spin measurement Flavor changing neutral current: t c(u)h Very good indicator of new physics Select tt events with one top in fully hadronic or lepton channel Use H γγ search selection Br(t c(u)h)<.83% (.53% expected) at 95% confidence T P ( =.69 Wednesday, August 4, 3

13 Conclusion How compatible is the new boson with the Standard Model? Measured properties all compatible with the SM Higgs: Combined signal strengths across all decay modes and also for the different production modes are compatible with SM production Couplings do not deviate from the SM predictions. Custodial symmetry is preserved is consistent with spin hypothesis No strong sign of ΓBSM Starting to probe differential signal strengths and directly search for new physics (Flavor Changing Neutral Current) All of the above measurements will be much more precise at 4TeV with more data and also smaller theoretical uncertainties REFERENCE: HIG3-5, ATLAS-CONF-34,CMS Public Note 3/, ATLAS-3-7, ATLAS-3-8 PUBLIC TWIKI: CMS ATLAS 3 Wednesday, August 4, 3

14 Additional Material 4 Wednesday, August 4, 3

15 Test Statistic q µ = log L(µ,θ µ) L(ˆµ,ˆθ) CL s = p s ps p b L hypothesis µ, θµ model of uncertainty Denominator maximized Likelihood -pb CLs quantifies the significance of the value (consistent with a fluctuating background? Or an excess consistent with signal hypothesis Example of Signal Background hypothesis testing: Red psuedo-data ( statistical toys) of signal background (expected signal predicted by SM cross-section) Blue psuedo-data ( statistical toys) background only Observed value is the value that minimizes the the above ratio in data (value of µ,θ most compatible with the data known as the best fit values) Distribution of test statistics follows a χ distribution: p-value obtained by integrating from the obs value to inf. Used to compute the confidence interval ± / SM..5 CMS Preliminary s = 7 TeV, L $ 5. fb s = 8 TeV, L $ 9.6 fb H # "" H # ZZ Combined H # "" H # ZZ Can include more than one hypothesis value (increase ndof in the chi). Here there are two hypotheses variables included in the likelihood: mass, σobs /σsm (signal strength based on SM cross-section) CLs corresponding to 68% is the contour around the best fit values (cross) m X (GeV) Wednesday, August 4, 3

16 4TEV PROJECTIONS Based on SNOWMASS studies: Extrapolate from current dataset to 3/fb (3/fb) at 4TeV with the present level of detector performance Scenarios for projected uncertainties: SCENARIO : all systematic uncertainties are left unchanged SCENARIO : Theoretical uncertainties scale by / and other systematic uncertainties scale by /sqrt(luminosity) (more optimistic scenario) 6 Wednesday, August 4, 3

17 Higgs Doublets In two Higgs Doublet models the yukawa couplings of fermions to neutral Higgs can be substantially modified MSSM check u,d coupling ratio Also in more general scenarios s = 7 TeV, L # leptons can virtually decouple 5. s from fb du, $ u, $ V the Higgs so test lepton/quark coupling ratio one is within the 68% CL for both Wednesday, August 4, 3 7 = 8 TeV, L # 9.6 fb - " ln L CMS Preliminary s = 7 TeV, L # 5. fb s = 8 TeV, L # 9.6 fb du, $ u, $ V Observed Exp. for SM H " ln L du CMS Preliminary s = 7 TeV, L # 5. fb s = 8 TeV, L # 9.6 fb lq, $ q, $ V Observed Exp. for SM H. 3 Excluded Figure : (Left) Likelihood scan versus ratio of couplings to down/up fermions λ du, with the two other free coupling modifiers, κ V and κ u, profiled together with all other nuisance parameters. (Right) Likelihood scan versus ratio of couplings to leptons and quarks λ, with lq

18 SPIN in CS Frame CMS Preliminary s = 8 TeV, L = 9.6 fb "/" SM X$## X$## X$## X$## m m m Observed (%gg) (%qq) (5%gg,5%qq) cos(*) 7: The SM extracted signal yield as a function of cos for the exp 8 Wednesday, August 4, 3

19 Differential Mu Agreement at low statistics is fair: P T" # " cos $ " χ /ndof p-value 6.9/ / /.64 N Jet "# JJ χ /ndof p-value 4.6/ / Wednesday, August 4, 3