H Workshop on Photon Physics and Simulation at Hadron Colliders - - PowerPoint PPT Presentation

▶
h
SMART_READER_LITE
LIVE PREVIEW

H Workshop on Photon Physics and Simulation at Hadron Colliders - - PowerPoint PPT Presentation

Jan 06, 2024 146 likes •579 views

H Workshop on Photon Physics and Simulation at Hadron Colliders 2019 Ruggero Turra on behalf of the ATLAS and CMS Collaboration INFN Milano 7 May 2019 Table of contents 1 Higgs boson and its properties 2 ATLAS and CMS analyses 3

slide-1
SLIDE 1

H → γγ

Workshop on Photon Physics and Simulation at Hadron Colliders 2019 Ruggero Turra

  • n behalf of the ATLAS and CMS Collaboration

INFN Milano

7 May 2019

slide-2
SLIDE 2

Table of contents

1 Higgs boson and its properties 2 ATLAS and CMS analyses 3 Conclusions

R.Turra (INFN Milano) H → γγ 7 May 2019 2 / 39

slide-3
SLIDE 3

Recent H → γγ analyses

Date1 Collaboration Topic Luminosity link 2018/6 ATLAS Mass 36 fb−1

  • Phys. Lett. B 784 (2018) 345

2017/11 ATLAS EFT from STXS 36 fb−1 ATL-PHYS-PUB-2017-018 2018/2 ATLAS STXS + fiducial/diff 36 fb−1

  • Phys. Rev. D 98 (2018) 052005

2018/7 ATLAS STXS + fiducial/diff 80 fb−1 ATLAS-CONF-2018-028 2018/4 CMS coupling 36 fb−1 JHEP 11 (2018) 185 2018/7 CMS fiducial/diff 36 fb−1 JHEP 01 (2019) 183 2019/3 CMS STXS (ggF and VBF) 77 fb−1 CMS-PAS-HIG-18-029 2019/4 ATLAS ttH 140 fb−1 ATLAS-CONF-2019-004 2018/10 CMS ttH 77 fb−1 CMS-PAS-HIG-18-018

1of the preprint R.Turra (INFN Milano) H → γγ 7 May 2019 3 / 39

slide-4
SLIDE 4

Section 1 Higgs boson and its properties

slide-5
SLIDE 5

Higgs properties

Excess compatible with Higgs boson firmly established by ATLAS+CMS in 2012. Measurements Mass: mH known at 0.2% (single experiment) σ × Br: inclusive, for each production-mode, fiducial region (STXS) (very

  • ptimized on the SM, acceptance extrapolations, model dependent)

Fiducial cross sections or differential cross sections in fiducial regions (minimal model dependence) Interpretations Spin and parity: 0+, other models excluded in Run 1. Signal strengths: µi = σi/σSM

i

(inclusive, per-production-mode, . . . ) Coupling modifiers to SM particles (k-framework) EFT interpretations, CP, . . .

R.Turra (INFN Milano) H → γγ 7 May 2019 5 / 39

slide-6
SLIDE 6

Higgs properties

Excess compatible with Higgs boson firmly established by ATLAS+CMS in 2012. Measurements Mass: mH known at 0.2% (single experiment) σ × Br: inclusive, for each production-mode, fiducial region (STXS) (very

  • ptimized on the SM, acceptance extrapolations, model dependent)

Fiducial cross sections or differential cross sections in fiducial regions (minimal model dependence) Interpretations Spin and parity: 0+, other models excluded in Run 1. Signal strengths: µi = σi/σSM

i

(inclusive, per-production-mode, . . . ) Coupling modifiers to SM particles (k-framework) EFT interpretations, CP, . . .

R.Turra (INFN Milano) H → γγ 7 May 2019 5 / 39

slide-7
SLIDE 7

Simplified template cross sections (STXS) stage 1

Yellow Report 4 - CERN-2017-002-M

Exclusive fiducial regions defined by production mode, pH

T , Nj, VBF-topology, pj1 T ,

pHjj

T , pV T

STXS bins

interpr.

Design measurement to split events according to STXS

R.Turra (INFN Milano) H → γγ 7 May 2019 6 / 39

slide-8
SLIDE 8

H → γγ

Small Br: 2.27 × 10−3 Loop decay: sensitive to BSM Simple final state: good resolution (1.4–2.1 GeV), good efficiency (≃ 40%) Very large background q¯ q/gg → γγ and fakes Falling background, can be modelled fitting data mγγ sidebands

R.Turra (INFN Milano) H → γγ 7 May 2019 7 / 39

slide-9
SLIDE 9

Section 2 ATLAS and CMS analyses

slide-10
SLIDE 10

General strategy for H → γγ

Shape (mγγ) analysis in categories of selected events Use fit to mγγ to extract signal and bkg yields in selected sample Modeling background mγγ distribution with analytical functions Signal shape from MC, as double sided Crystall Ball (ATLAS) or sum of Gaussians (CMS) Photons are selected using shower shapes and isolation: ATLAS: rectangular cuts (tight selection) CMS: BDT, usually used as a continous variable Extract signal in different categories pure of events under study (production mode, STXS, kinematic bin, . . . ) Categories defined from properies of selected objects (kinematic, identification, quality, . . . ) CMS has a predictor of the expected resolution

R.Turra (INFN Milano) H → γγ 7 May 2019 9 / 39

slide-11
SLIDE 11

Coupling categorization in ATLAS (STXS)

ATLAS-CONF-2018-028

To measure many cross-sections with small correlation split events in pure categories

ggF ttH

prod modes

VH VBF

29 reco-categories inspired by STXS

R.Turra (INFN Milano) H → γγ 7 May 2019 10 / 39

slide-12
SLIDE 12

Categorization in CMS (ggF and VBF STXS)

CMS-PAS-HIG-18-029

Diphoton-BDT using photon kinematic, BDT-id score, mass resolution, vertex probability VBF categorization: dijet-BDT trained with signal: VBF, background: ggF+jets and non-Higgs (from control region inverting photon id score) 6 categories using dijet ⊗ diphoton-BDT and pHjj

T , mjj, pj1 T mimic STXS (2J, 3J,

BSM, rest) ggF categorization: mimic STXS using pγγ

T

and number of jets. Split also by diphoton BDT (“Tag”). All BDT validated on Z → ee

R.Turra (INFN Milano) H → γγ 7 May 2019 11 / 39

slide-13
SLIDE 13

Categorization in CMS (ggF and VBF STXS)

Category signal composition (%) 20 40 60 80 100

73 11 5 1 1 1 3 4 78 10 4 1 1 1 2 3 83 8 2 1 2 3 27 55 1 6 1 1 5 5 27 54 2 5 1 1 1 5 5 1 1 63 1 14 1 1 3 1 8 6 2 2 65 12 1 1 2 1 8 6 51 1 15 1 2 6 2 10 1 9 2 49 17 2 2 5 2 10 9 1 48 18 2 1 2 14 12 16 19 41 1 1 2 1 1 5 13 15 19 39 1 1 2 1 1 6 1 14 14 59 2 2 1 1 5 1 15 1 16 1 54 1 2 1 1 6 1 16 7 59 2 2 1 1 5 1 1 20 7 1 56 1 1 3 1 1 5 1 1 22 4 1 62 1 2 5 1 24 3 2 56 1 1 2 5 1 30 1 3 1 1 1 1 1 13 3 58 7 9 1 1 2 3 4 1 2 4 2 1 18 5 40 4 13 2 1 2 1 1 1 3 7 5 15 26 26 7 2 3 1 3 5 2 3 7 4 3 6 21 13 16 8 1 6 1 2 3 2 8 16 18 8 1 2 2 1 26 7 2 2 37 2 2 5 3 1 39 7

STXS process

ggH 0J ggH 1J low ggH 1J med ggH 1J high ggH 1J BSM ggH 2J low ggH 2J med ggH 2J high ggH 2J BSM ggH VBF-like 2J ggH VBF-like 3J VBF 2J-like VBF 3J-like VBF rest VBF BSM VBF VH-like Other

Event category

0J Tag0 0J Tag1 0J Tag2 1J low Tag0 1J low Tag1 1J med Tag0 1J med Tag1 1J high Tag0 1J high Tag1 1J BSM 2J low Tag0 2J low Tag1 2J med Tag0 2J med Tag1 2J high Tag0 2J high Tag1 2J BSM Tag0 2J BSM Tag1 VBF 2J-like Tag0 VBF 2J-like Tag1 VBF 3J-like Tag0 VBF 3J-like Tag1 VBF rest VBF BSM

γ γ → H Simulation Preliminary CMS

13 TeV (2017)

R.Turra (INFN Milano) H → γγ 7 May 2019 12 / 39

slide-14
SLIDE 14

ttH-categories

ATLAS-CONF-2019-004

ATLAS, great improvement using agressive optimization: Leptonic (≥ 1ℓ, ≥ 1b), hadronic (≥ 1b, ≥ 2j, 0ℓ) regions Train two BDT with low-level variables: pT/mγγ, η, φ of photons, pµ of up to two leptons, pµ of up to four/six jets (lep/had, pT sorted), magnitude and φ of ET-miss Categories defined from the BDT output Trained on MC ttH signal and control regions (non-tight non-isolated) for background CMS: Leptonic (≥ 1ℓ, ≥ 1b), hadronic (≥ 2j, 0ℓ) regions Train two BDT with: pT/mγγ, η, BDT-id of photons, ∆φγγ or φ, pT and η of diphoton (had only), number of (b)-jets, pT and η of the first three (four) jets (pT

  • rdering),

all jets pT (had), b-discriminant, pT and η of the lepton, ET-miss

Categories defined from the numbers of leptons (1/2) and BDT output Trained on MC ttH signal, bkg MC, ggF+VBF MC (only had)

R.Turra (INFN Milano) H → γγ 7 May 2019 13 / 39

slide-15
SLIDE 15

ttH-categories Main problem: theoretical systematic for ggF contamination (ggF+HF) ATLAS: 100% uncertainty on ggF, VBF, VH production (supported by H → 4ℓ, ttbb, Vb). Impact 3-4% CMS (only ggF): parton shower (from difference in jet multiplicity aMCNLO/data tt+j), gluon splitting (scaling the fraction of events from ggF+b in simulation by the measured difference data/simulation

  • f σ(ttbb)/σ(ttjj)). Impact 2%.

Better recipe?

R.Turra (INFN Milano) H → γγ 7 May 2019 14 / 39

slide-16
SLIDE 16

Background modeling in ATLAS

Background mγγ distributions are fitted directly on data with simple analytical functions both in ATLAS and CMS Different approach how the functional form (exponential, . . . ) are selected In ATLAS, for each category, one functional form is selected by dedicated studies on MC or control regions Due to the little s/b, small mismodeling on the shape can bias the signal yield Quantify bias using closure tests injecting 0 signal events: spurious signal Select functional form which pass criteria based on the spurious signal

R.Turra (INFN Milano) H → γγ 7 May 2019 15 / 39

slide-17
SLIDE 17

Spurious signal (ATLAS)

Build a background-only template, usually mixing γγ from Sherpa NLO and γj from control region Run a signal + background fit The spurious signal (the bias) is the number of fitted events (positive or negative) Assume the systematic on the signal yield to be the maxmH |bias| changing mH in a window Problems: MC is not data: need to try different variations (e.g. γγ purity) MC is limited: statistical fluctuation in the MC can increase the estimated spurious signal Limiting factor of the procedure: not scalable Produce faster MC (e.g. smeared truth MC, LO generators): bilions of events Use more flexible function (even not analytic) so that you expect better modeling (how to validate?) Remove statistical fluctuation from your template

R.Turra (INFN Milano) H → γγ 7 May 2019 16 / 39

slide-18
SLIDE 18

Background modeling in CMS (discrete profiling)

arxiv 1408.6865

Let the s + b fit on data choose the best background function and profile Consider functions with different number of degrees of freedom from exponential, power-law, . . . families: select functions that can describe data with injection test and F-test Add penalty term to the likelihood to account for the number of free parameters NB −2 log ˜ L = −2 log L + NB

(GeV)

γ γ

m

110 115 120 125 130 135 140 145 150

Events / GeV

50 100 150 200 250

Laurent Exponential Power Law Polynomial µ

  • 1
  • 0.5

0.5 1 1.5 2 2.5

+ correction Λ

206 208 210 212 214 216 218 220 222

  • Approx. p-value

R.Turra (INFN Milano) H → γγ 7 May 2019 17 / 39

slide-19
SLIDE 19

STXS results with 80 fb−1 (ATLAS)

ATLAS-CONF-2018-028

2 − 1 − 1 2 3 4 5

SM

x B) σ x B) / ( σ (

Total Stat. Syst. SM Preliminary ATLAS

−1

= 13 TeV, 79.8 fb s | < 2.5

H

, |y γ γ → H ( ) Total Stat. Syst. Top )

− 0.19 + 0.23 − 0.34 + 0.37

(

− 0.38 + 0.44

1.13 VH, leptonic )

− 0.25 + 0.29 − 0.59 + 0.65

(

− 0.64 + 0.71

1.38 Hqq, BSM−like → ggF + qq 0.23 ) ±

− 0.43 + 0.45

(

− 0.49 + 0.50

0.76 <200 GeV

j T

Hqq, 0<p → qq )

− 0.21 + 0.30 − 0.34 + 0.36

(

− 0.40 + 0.47

1.40 ggF, >= 2j )

− 0.21 + 0.29

0.47 ± (

− 0.52 + 0.56

0.65 <200 GeV

H T

ggF, 1j, 120<p )

− 0.35 + 0.49 − 0.68 + 0.70

(

− 0.76 + 0.85

1.51 <120 GeV

H T

ggF, 1j, 60<p )

− 0.21 + 0.27 − 0.42 + 0.43

(

− 0.47 + 0.50

0.89 <60 GeV

H T

ggF, 1j, 0<p )

− 0.31 + 0.43

0.52 ± (

− 0.61 + 0.68

1.23 ggF, 0j )

− 0.14 + 0.16

0.17 ± (

− 0.22 + 0.23

0.92

Difficult to separate

✞ ✝ ☎ ✆

ggF 0j/ggF 1j pT < 60 GeV and

✞ ✝ ☎ ✆

qq → Hqq/ggF 2j Interpretation µ = 1.06 ± 0.08(stat)+0.08

−0.07(exp)+0.07 −0.06(theo)

Main theoretical uncertainty on µ: PS and UE (Pythia vs Herwig or AZNLO eigentunes variations for ggF); renormalization, factorization scales

R.Turra (INFN Milano) H → γγ 7 May 2019 18 / 39

slide-20
SLIDE 20

STXS (ggF,VBF) with 77 fb−1 (CMS)

CMS-PAS-HIG-18-029

theo

σ /

proc

σ

2 − 2 4 6 8

  • 0.3

+0.4

0.8

qqH

  • 0.8

+0.8

2.2

ggH BSM

  • 0.5

+0.6

0.8

ggH GE2J

  • 0.7

+0.9

1.7

ggH 1J high

  • 0.4

+0.4

0.7

ggH 1J med

  • 0.5

+0.6

1.3

ggH 1J low

  • 0.20

+0.20

1.18

ggH 0J

profiled

H

m Observation SM Prediction

Preliminary

CMS γ γ → H TeV) (13

  • 1

77.4 fb

1.00

  • 0.28
  • 0.04

0.08 0.08 0.07

  • 0.02
  • 0.28

1.00 0.08 0.06 0.01 0.09

  • 0.07
  • 0.04

0.08 1.00 0.10

  • 0.08

0.21

  • 0.15

0.08 0.06 0.10 1.00

  • 0.02

0.18

  • 0.18

0.08 0.01

  • 0.08
  • 0.02

1.00 0.32

  • 0.45

0.07 0.09 0.21 0.18 0.32 1.00

  • 0.35
  • 0.02
  • 0.07
  • 0.15
  • 0.18
  • 0.45
  • 0.35

1.00

ggH 0J ggH 1J low ggH 1J med ggH 1J high ggH 2J ggH BSM qqH ggH 0J ggH 1J low ggH 1J med ggH 1J high ggH 2J ggH BSM qqH

1 − 0.8 − 0.6 − 0.4 − 0.2 − 0.2 0.4 0.6 0.8 1 γ γ → H Supplementary CMS (13 TeV)

  • 1

77.4 fb

Less correlation than ATLAS

✞ ✝ ☎ ✆

ggF 0j/ggF 1j pT < 60 GeV (-28% vs -48%) ATLAS/CMS similar for

✞ ✝ ☎ ✆

qq → Hqq/ggF 2j ATLAS decided to merge BSM STXS bins to avoid large correlations

R.Turra (INFN Milano) H → γγ 7 May 2019 19 / 39

slide-21
SLIDE 21

ttH with 41.5 fb−1 (CMS)

CMS-PAS-HIG-18-018 S/(S+B) Weighted Events / GeV

5 10 15 20

Data S+B Background σ 1 ± σ 2 ± S/(S+B) weighted All categories =1.3

H t t

µ

  • Preliminary

CMS TeV) (13

  • 1

41.5 fb H t t γ γ → H (GeV)

γ γ

m

100 110 120 130 140 150 160 170 180 5 − 5 10

B component subtracted

10 20 30 40 50 60 70 80 90 100 0.5 1 1.5 2 2.5 3 3.5 4 H t t bbH tHq tHW ggH VBF WH leptonic ZH leptonic WH hadronic ZH hadronic 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 eff

σ FWHM

0.1 0.2 0.3 0.4 0.5 0.6 0.7

S/(S+B)

ttH Hadronic 0 2.4 expected events ttH Hadronic 1 3.3 expected events ttH Hadronic 2 5.2 expected events ttH Leptonic 0 2.7 expected events ttH Leptonic 1 1.2 expected events

Signal fraction (%) Width (GeV)

eff

σ ± S/(S+B) in

γ γ → H Preliminary CMS (13 TeV)

  • 1

41.5 fb

Very pure categories. Contamination from tH, ggF (hadronic categories), VH (leptonic categories) In the best category: s/(s+b)=70%. ttH/all-Higgs = 89%. Resolution σ68 = 1.66 GeV.

R.Turra (INFN Milano) H → γγ 7 May 2019 20 / 39

slide-22
SLIDE 22

ttH with 140 fb−1 (ATLAS)

  • Phys. Lett. B 784 (2018) 173, ATLAS-CONF-2019-004, CMS-HIG-17-035

110 120 130 140 150 160 [GeV]

γ γ

m 5 10 15 20 25 30 Sum of Weights / 1.375 GeV

Data Continuum Background Total Background Signal + Background

Preliminary ATLAS

  • 1

= 13 TeV, 139 fb s

= 125.09 GeV

H

m All categories ln(1+S/B) weighted sum

20 40 60 80 100 120 140 160 180 Events

Data =1.4) µ H ( t t H Higgs t Non-t

  • Cont. Bkg.

Had categories Lep categories Preliminary ATLAS

  • 1

=13 TeV, 139 fb s Had 4 Had 3 Had 2 Had 1 Lep 3 Lep 2 Lep 1 20 Data - Bkg. =1.4) µ H ( t t

In the best category: s/(s+b)=60%. ttH/all-Higgs = 95%. Resolution σ68 = 1.39 GeV. Luminosity [fb−1]

  • bserved

expected ATLAS γγ 140 4.9σ 4.2σ CMS γγ 41.5 3.1σ 2.2σ CMS γγ 41.5+35.9 4.1σ 2.7σ ATLAS combination 80 5.8σ 4.9σ ATLAS combination 80 + Run1 6.3σ 5.1σ CMS combination 36 + Run1 5.2σ 4.2σ

R.Turra (INFN Milano) H → γγ 7 May 2019 21 / 39

slide-23
SLIDE 23

Strategy for fiducial and differential cross sections

JHEP 01 (2019) 183, , ATLAS-CONF-2018-028

Define fiducial regions close to experimental analysis cuts Avoid selections that can distort observables For each reco-bin fit mγγ distribution In addition CMS splits events by expected resolution CMS uses likelihood with matrix method (no regularization, since bins are large) ATLAS uses bin-by-bin unfolding (bias negligible) Inclusive fiducial “diphoton” region ATLAS CMS |ηγ| < 1.37 or 1.52 < |ηγ| < 2.37 |ηγ| < 2.5 piso

T (R = 0.2)/pγ T < 0.05

piso

T (R = 0.3) < 10 GeV

pγ1

T /mγγ > 0.35, pγ2 T /mγγ > 0.25

pγ1

T /mγγ > 1/3, pγ2 T /mγγ > 1/4

R.Turra (INFN Milano) H → γγ 7 May 2019 22 / 39

slide-24
SLIDE 24

Fiducial cross sections (ATLAS and CMS 36 fb−1)

(fb)

fid

σ

1 −

10 1 10

2

10

3

10 1-b-jet ≥ 1-lepton, ≥

miss T

1-lepton, high-p

miss T

1-lepton, low-p >1

b

N =1

b

N >1

lepton

N =1

lepton

N > 200 GeV

miss T

p 200 GeV ≤

miss T

100 < p Inclusive

fb

  • 0.16

+0.22

  • 0.04

fb

  • 0.20

+0.37

0.22 fb

  • 1.4

+1.4

0.3 fb

  • 1.5

+1.8

1.8 fb

  • 3.7

+3.6

2.2 fb

  • 0.23

+0.37

0.20 fb

  • 1.3

+1.4

0.5 fb

  • 0.09

+0.14

0.03 fb

  • 0.46

+0.53

0.36 13 fb ± 84 fb

  • 0.16

+0.22

  • 0.04

fb

  • 0.20

+0.37

0.22 fb

  • 1.4

+1.4

0.3 fb

  • 1.5

+1.8

1.8 fb

  • 3.7

+3.6

2.2 fb

  • 0.23

+0.37

0.20 fb

  • 1.3

+1.4

0.5 fb

  • 0.09

+0.14

0.03 fb

  • 0.46

+0.53

0.36 13 fb ± 84 fb

  • 0.16

+0.22

  • 0.04

fb

  • 0.20

+0.37

0.22 fb

  • 1.4

+1.4

0.3 fb

  • 1.5

+1.8

1.8 fb

  • 3.7

+3.6

2.2 fb

  • 0.23

+0.37

0.20 fb

  • 1.3

+1.4

0.5 fb

  • 0.09

+0.14

0.03 fb

  • 0.46

+0.53

0.36 13 fb ± 84

NNLOPS aMC@NLO, syst unc. ⊕ Data, stat Systematic uncertainty CYRM-2017-002

) from

γ γ → H

(

SM

σ

CMS

(13 TeV)

  • 1

35.9 fb

1 −

10 × 2 1 2 3 4 56 10 20 30

2

10 [fb]

fid

σ

ttH-enhanced

miss T

E High 1 ≥

lepton

N VBF-enhanced Diphoton fiducial

95% C.L. 95% C.L. 95% C.L.

ATLAS

  • 1

= 13 TeV, 36.1 fb s , γ γ → H Data, tot. unc.

  • Syst. unc.

= 125.09 GeV

H

m XH LO +

3

N XH default MC + H → gg XH Powheg NNLOPS + bbH + ttH + VH = VBF+ XH

Define fiducial regions where some production are enhanced Don’t try to split production modes with additional variables Would be good to harmonize on the most interesting regions (which ones?)

R.Turra (INFN Milano) H → γγ 7 May 2019 23 / 39

slide-25
SLIDE 25

Differential cross sections

Define fiducial regions only with kinematic cuts, close to the ones used in the analysis: small model dependency Many variables are investigated2: pγγ

T , |yγγ|, Nj, pj1 T , pj2 T , |yj1|, |yj2|, | cos(θ∗)|, ∆φjj,

|∆yjj|, ∆φγγ,jj, |∆φγγ,j1|, |∆yγγ,j1|, mjj, |ηjj − ηjj|, |∆ηjj|, pmiss

T

, Nb−jets, Nℓ. Anything missing? Interesting to use different jet definitions (pT, central/forward) ? Also double differential cross sections: pγγ

T × Nj, pγγ T × | cos(θ∗)|. More?

Already some variables are measured in specific phase space. Interesting to see differential distributions in specific phase space (e.g. pγγ

T , ∆φjj within different

fiducial regions)?

2ATLAS only, CMS only R.Turra (INFN Milano) H → γγ 7 May 2019 24 / 39

slide-26
SLIDE 26

CMS differential cross section (36 fb−1)

JHEP 01 (2019) 183

Unfolded (matrix method) distribution for pγγ

T , yγγ, pj1 T , Nb−jets

50 100 150 200 250 300 350 400 450 500

(fb/GeV)

γ γ T

p ∆ /

fid

σ ∆

4 −

10

3 −

10

2 −

10

1 −

10 1 10

2

10

3

10

4

10

5

10

MC@NLO H a t HX = VBF + VH + t + HX NNLOPS , MC@NLO ggH a + HX MC@NLO ggH a + HX POWHEG ggH CYRM-17-002 ) from γ γ → H ( SM σ syst unc. ⊕ Data, stat Systematic uncertainty >350 GeV)/150

γ γ T

(p

fid

σ

CMS (13 TeV)

  • 1

35.9 fb γ γ → H (GeV)

γ γ T

p

100 200 300 400 500 Ratio to prediction 0.5 − 0.5 1 1.5 2 2.5 0.5 1 1.5 2 2.5

| (fb)

γ γ

|y ∆ /

fid

σ ∆

10

2

10

3

10

MC@NLO H a t HX = VBF + VH + t + HX NNLOPS , MC@NLO ggH a + HX MC@NLO ggH a + HX POWHEG ggH CYRM-17-002 ) from γ γ → H ( SM σ syst unc. ⊕ Data, stat Systematic uncertainty

CMS (13 TeV)

  • 1

35.9 fb γ γ → H |

γ γ

|y

0.5 1 1.5 2 2.5 Ratio to prediction 1 1.5 2 50 100 150 200 250 300

(fb/GeV)

1

j T

p ∆ /

fid

σ ∆

2 −

10

1 −

10 1 10

MC@NLO H a t HX = VBF + VH + t + HX NNLOPS , MC@NLO ggH a + HX MC@NLO ggH a + HX POWHEG ggH CYRM-17-002 ) from γ γ → H ( SM σ syst unc. ⊕ Data, stat Systematic uncertainty >200 GeV)/80

1 j T

(p

fid

σ | < 2.5

1

j

η > 30 GeV, |

1

j T

p

CMS (13 TeV)

  • 1

35.9 fb γ γ → H (GeV)

1

j T

p

50 100 150 200 250 300 Ratio to prediction 1 − 1 2 3 4 1 >1

(fb)

b jet

N ∆ /

fid

σ ∆

1 10

2

10

3

10

4

10

MC@NLO H a t HX = VBF + VH + t + HX NNLOPS , MC@NLO ggH a + HX MC@NLO ggH a + HX POWHEG ggH CYRM-17-002 ) from γ γ → H ( SM σ syst unc. ⊕ Data, stat Systematic uncertainty | < 2.5

jet

η > 30 GeV, |

jet T

p

CMS (13 TeV)

  • 1

35.9 fb γ γ → H

b jet

N

1 >1 Ratio to prediction 2 − 2 4 6 8

R.Turra (INFN Milano) H → γγ 7 May 2019 25 / 39

slide-27
SLIDE 27

ATLAS differential cross sections (80 fb−1)

ATLAS-CONF-2018-028, JHEP 01 (2019) 183

Unfolded (bin by bin) distribution for pγγ

T , yγγ, pj1 T , Nb−jets (lepton-veto)

50 100 150 200 250 300 350

2 −

10

1 −

10 1 [fb/GeV]

γ γ T

p / d

fid

σ d Preliminary ATLAS

  • 1

= 13 TeV, 79.8 fb s , γ γ → H Data, tot. unc.

  • Syst. unc.

XH default MC + H → gg bbH + ttH + VH = VBF+ XH XH SCET + ⊕ NNLOJET 50 100 150 200 250 300 350 [GeV]

γ γ T

p 1 2 XH Ratio to default MC +

1 2

20 40 60 80 | [fb]

γ γ

y / d|

fid

σ d Preliminary ATLAS

  • 1

= 13 TeV, 79.8 fb s , γ γ → H data, tot. unc.

  • syst. unc.

XH default MC + H → gg bbH + ttH + VH = VBF+ XH XH SCETlib+MCFM8 + H → gg 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 |

γ γ

y | 1 2 XH Ratio to default MC +

200

2 −

10

1 −

10 1 [fb/GeV]

j1 T

p / d

fid

σ d Preliminary ATLAS

  • 1

= 13 TeV, 79.8 fb s , γ γ → H data, tot. unc.

  • syst. unc.

XH default MC + H → gg bbH + ttH + VH = VBF+ XH XH NNLOJET + H → gg XH SCETlib (STWZ) + H → gg ≥

jets

N = 0.4, R

t

k anti 50 100 150 200 250 300 350 [GeV]

j1 T

p 1 2 XH Ratio to default MC +

1 2 3 4

2 −

10

1 −

10 1 10

2

10

3

10 [fb]

fid

σ Preliminary ATLAS

  • 1

= 13 TeV, 79.8 fb s , γ γ → H data, tot. unc.

  • syst. unc.

XH default MC + H → gg bbH + ttH + VH = VBF+ XH ttH = 0

µ e+

N 1, ≥ | < 2.5) η > 30 GeV, |

T

(p

jets

N = 0 = 1 2 ≥ b-jets

N 1 2 XH Ratio to default MC +

p-value(χ2) data/SM > 30% EFT interpretation: introduce additional CP-even and CP-odd interactions in SILH framework inputs: pγγ

T , Nj, mjj, |∆φjj|, pj1 T and

their correlations

HW

HB HW HW HB

c

c = c c = c

HW

c ~

~ ~

0.2 0.1 0.1 0.2 0.4 0.2 0.2 0.4

contours hold 68%, 95% CL γ γ

σ σ Δ

− − − −

⊕ ⊕ ⊕

= 125.09 GeV

H

, m

  • 1

= 13 TeV, 36.1 fb s , γ γ → H Standard Model 68% CL 95% CL

R u n 1 9 5 % C L R u n 1 6 8 % C L

ATLAS

R.Turra (INFN Milano) H → γγ 7 May 2019 26 / 39

slide-28
SLIDE 28

Mass measurement γγ + 4ℓ with 36 fb−1 (ATLAS)

  • Phys. Lett. B 784 (2018) 345

Hundreds of systematics on the energy/momentum scale Energy scale mostly from Z → ℓℓ (

  • pℓ

T

  • ≃ 40 GeV) comparing

data/MC

H → γγ (pγ

T ≃ 60 GeV) starts to be dominated by systematics

123 124 125 126 127 128 [GeV]

H

m

Total

  • Stat. only

ATLAS

Total (Stat. only)

Run 1 ATLAS + CMS

0.21) GeV ± 0.24 ( ± 125.09

Combined Run 1+2

0.16) GeV ± 0.24 ( ± 124.97

Combined Run 2

0.18) GeV ± 0.27 ( ± 124.86

Combined Run 1

0.37) GeV ± 0.41 ( ± 125.38

γ γ → H Run 1+2

0.19) GeV ± 0.35 ( ± 125.32

l 4 → H Run 1+2

0.30) GeV ± 0.30 ( ± 124.71

γ γ → H Run 2

0.21) GeV ± 0.40 ( ± 124.93

l 4 → H Run 2

0.36) GeV ± 0.37 ( ± 124.79

γ γ → H Run 1

0.43) GeV ± 0.51 ( ± 126.02

l 4 → H Run 1

0.52) GeV ± 0.52 ( ± 124.51

  • 1

= 13 TeV, 36.1 fb s : Run 2 ,

  • 1

= 7-8 TeV, 25 fb s : Run 1

Any interest to improve mH→γγ with sys 100 MeV?

R.Turra (INFN Milano) H → γγ 7 May 2019 27 / 39

slide-29
SLIDE 29

Section 3 Conclusions

slide-30
SLIDE 30

Conclusions Precision studies have been presented: no deviation from SM Two complementary approaches: coupling/STXS (very optimized, model dependent) vs fiducial/differential cross sections How much differential we should be? As much as possible? Or not useful to quote results with very large errors and correlations? In STXS we have tens of cross sections. What do we want to optimize? How to share results? Just values and covariance? Main difficulty from the experimental point of view: modeling the shape of the background and its systematic Being more differential means more difficult to evaluate theoretical uncertainties (e.g. ggF+HF) Now working on final Run2 papers with 140 fb−1. More complicated interpretations (EFT)

R.Turra (INFN Milano) H → γγ 7 May 2019 29 / 39

slide-31
SLIDE 31
slide-32
SLIDE 32

Section 4 Backup

slide-33
SLIDE 33

Migrations ATLAS

R.Turra (INFN Milano) H → γγ 7 May 2019 31 / 39

slide-34
SLIDE 34

Migrations ATLAS (no ggF categories)

R.Turra (INFN Milano) H → γγ 7 May 2019 32 / 39

slide-35
SLIDE 35

Systematic on mH from H → γγ (ATLAS)

Source Systematic uncertainty on mγγ

H [MeV]

EM calorimeter cell non-linearity ±180 EM calorimeter layer calibration ±170 Non-ID material ±120 ID material ±110 Lateral shower shape ±110 Z → ee calibration ±80 Conversion reconstruction ±50 Background model ±50 Selection of the diphoton production vertex ±40 Resolution ±20 Signal model ±20

R.Turra (INFN Milano) H → γγ 7 May 2019 33 / 39

slide-36
SLIDE 36

Signal shape (ATLAS)

[GeV]

γ γ

m 115 120 125 130 135 140 / 0.5 GeV

γ γ

m 1/N dN/d 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

Simulation ATLAS = 13 TeV s = 125 GeV

H

, m γ γ → H

=1.59 GeV)

68

σ ggH 0J Cen ( MC Signal Model =2.10 GeV)

68

σ ggH 0J Fwd ( MC Signal Model

R.Turra (INFN Milano) H → γγ 7 May 2019 34 / 39

slide-37
SLIDE 37

STXS merged scheme (ATLAS)

All Higgs events, |yH | < 2.5 ggF + gg!Z(!qq)H, 0-jet ggF 0J Fwd, Cen (28, 29) 1-jet, pH

T < 60 GeV

ggF 1J Low (27) 60 ≤ pH

T < 120 GeV

ggF 1J Med (26) 120 ≤ pH

T < 200 GeV

ggF 1J High (25) pH

T > 200 GeV

ggF 1J BSM (24) ≥ 2-jet, not VBF-like, pH

T > 200 GeV

ggF 2J BSM (20) pH

T < 60 GeV

ggF 2J Low (23) 60 ≤ pH

T < 120 GeV

ggF 2J Med (22) 120 ≤ pH

T < 200 GeV

ggF 2J High (21) VBF-like*, pH j j

T

< 25 GeV pH j j

T

≥ 25 GeV qq0!Hqq0 (VBF + V H hadronic), pj

T < 200 GeV,

VBF-like*, pH j j

T

< 25 GeV VBF low-pH j j

T

BDT tight, loose (18, 19) pH j j

T

≥ 25 GeV VBF high-pH j j

T

BDT tight, loose (16, 17) V H-like† VH had BDT tight, loose (14, 15) Rest pj

T > 200 GeV

qqH BSM (13) V H (leptonic decays), q ¯ q ! W H VH lep High, Low (9, 10) q ¯ q ! ZH, gg ! ZH (Z!νν) VH MET High, Low (11, 12) (Z!``) VH dilep (8) top (t¯ tH, tHq, tHW) (had decays) ttH had BDT1-4 (4-7) (lep decays) ttH lep BDT1-3 (1-3) b¯ bH (merged at all stages with ggF)

BSM-like BSM-like ggF, ≥ 2 jet

Reconstruction Categories STXS Regions

*VBF-like: mj j > 400 GeV, |∆yj j| > 2.8

†V H-like: 60 < mj j < 120 GeV

R.Turra (INFN Milano) H → γγ 7 May 2019 35 / 39

slide-38
SLIDE 38

ATLAS STXS reco categories

R.Turra (INFN Milano) H → γγ 7 May 2019 36 / 39

slide-39
SLIDE 39

ATLAS STXS efficiencies

91 90 38 3 24 1 3 8 2 5 5 5 51 1 25 5 12 1 2 1 76 2 24 6 1 10 8 8 2 4 1 70 1 10 1 2 4 3 3 69 3 4 3 34 2 4 2 1 1 8 56 1 3 9 4 26 9 3 6 3 1 10 71 2 1 11 4 25 30 10 3 8 4 1 11 79 2 3 8 5 14 43 11 7 4 2 26 14 4 5 1 1 4 3 18 30 3 1 1 36 68 3 8 2 4 5 18 3 2 2 2 20 35 3 1 4 8 10 2 7 10 11 9 10 4 6 4 5 3 5 3 1 12 23 3 2 1 1 22 25 69 77 5 2 44 49 6 4 81 2 15 18 1 1 18 1 1 1 2 1 1 4 5 9 16 1 54 75 86 91 76 90 95 2 2 1 6 4 2 1 13 5 1 1 1 2 2 3 2 2 2 1 1 1 1 1 ggF 0J Cen ggF 0J Fwd ggF 1J Low ggF 1J Med ggF 1J High ggF 1J BSM ggF 2J Low ggF 2J Med ggF 2J High ggF 2J BSM BDT loose

Hjj T

VBF Low-p BDT tight

Hjj T

VBF Low-p BDT loose

Hjj T

VBF High-p BDT tight

Hjj T

VBF High-p VH had BDT loose VH had BDT tight qqH BSM VH MET Low VH MET High VH lep Low VH lep High VH dilep ttH had BDT4 ttH had BDT3 ttH had BDT2 ttH had BDT1 ttH lep BDT3 ttH lep BDT2 ttH lep BDT1

Reconstruction Category

ggF (0-jet) < 60 GeV)

H T

ggF (1-jet, p < 120 GeV)

H T

p ≤ ggF (1-jet, 60 < 200 GeV)

H T

p ≤ ggF (1-jet, 120 200 GeV) ≥

H T

ggF (1-jet, p < 60 GeV)

H T

2-jet, p ≥ ggF ( < 120 GeV)

H T

p ≤ 2-jet, 60 ≥ ggF ( < 200 GeV)

H T

p ≤ 2-jet, 120 ≥ ggF ( 200 GeV) ≥

H T

2-jet, p ≥ ggF ( ggF (VBF-like, 3-jet veto) ggF (VBF-like, 3-jet) Hqq (VBF-like 3-jet veto) → qq Hqq (VBF-like 3-jet) → qq Hqq (VH) → qq Hqq (rest) → qq 200 GeV) ≥

j T

Hqq (p → qq ν Hl → qq Hll → qq Hll → gg ttH tHq tWH bbH

STXS Region 10 20 30 40 50 60 70 80 90 100

Simulation Preliminary ATLAS 125.09 GeV =

H

= 13 TeV, m s , γ γ → H

Region purity / Category (%)

R.Turra (INFN Milano) H → γγ 7 May 2019 37 / 39

slide-40
SLIDE 40

ATLAS generators

R.Turra (INFN Milano) H → γγ 7 May 2019 38 / 39

slide-41
SLIDE 41

CMS STXS BDT validation

0.2 0.4 0.6 0.8 1 Dijet BDT score 1 10 Events/0.02

Data simulation

  • e

+

e → Z

  • syst. unc.

⊕ Simulation stat.

3

10 ×

Preliminary CMS

(13 TeV, 2017)

  • 1

41.5 fb 0.2 0.4 0.6 0.8 1 Diphoton BDT score 10

2

10

3

10

4

10 Events/0.02

Data simulation

  • e

+

e → Z

  • syst. unc.

⊕ Simulation stat.

3

10 ×

Preliminary CMS

(13 TeV, 2017)

  • 1

41.5 fb

R.Turra (INFN Milano) H → γγ 7 May 2019 39 / 39