Probing for anomalous couplings in production and decay 4 at CMS - - PowerPoint PPT Presentation

Probing for anomalous 𝐼𝑊𝑊 couplings in production and decay 𝐼 → 4ℓ at CMS

Heshy Roskes (Johns Hopkins University) for the CMS collaboration The 26th International Workshop on Weak Interactions and Neutrinos (WIN2017) UC Irvine June 20, 2017

Animated gifs on some slides. See ppt version or separate indico attachment.

1

• Search for anomalous 𝐼𝑊𝑊 couplings in production and

decay in the 𝐼 → 4𝑚 channel

• Kinematics of decay
• New: kinematics of jets from

VBF and 𝑊𝐼 production

• Use matrix element (MELA) discriminants
• optimally select VBF and 𝑊𝐼 events
• optimally separate different contributions to the amplitude
• Combine with Run 1 CMS analysis

Anomalous couplings

2

𝐼 125 → 4ℓ

3

• What is it?
• How does it interact with other particles?

References: Run 1:

• CMS-HIG-14-018

spin anomalous couplings Run 2:

• CMS-PAS-HIG-16-041

properties

• CMS-PAS-HIG-17-011

anomalous couplings

CMS&ATLAS results

CMS

• Study of the mass and spin-parity of the Higgs

boson candidate via its decays to Z boson pairs CMS-HIG-12-041, arXiv:1212.6639

• Measurement of the properties of a Higgs

boson in the four-lepton final state arXiv:1312.5353, CMS-HIG-13-002

• Constraints on the spin-parity and anomalous

HVV couplings of the Higgs boson in proton collisions at 7 and 8 TeV arXiv:1411.3441, CMS- HIG-14-018

• Limits on the Higgs boson lifetime and width

from its decay to four charged leptons arXiv:1507.06656, CMS-HIG-14-036

• Combined search for anomalous pseudoscalar

HVV couplings in VH production and H to VV decay arXiv:1602.04305, CMS-HIG-14-035

• Measurements of properties of the Higgs boson

and search for an additional resonance in the four-lepton final state at √s = 13 TeV, CMS-PAS- HIG-16-033

• Constraints on anomalous Higgs boson

couplings in production and decay H→4ℓ, CMS- PAS-HIG-17-011

ATLAS

• Evidence for the spin-0 nature of the Higgs

boson using ATLAS data ATLAS arXiv:1307.1432

• Study of the spin and parity of the Higgs boson

in diboson decays with the ATLAS detector ATLAS arXiv:1506.05669

• Test of CP Invariance in vector-boson fusion

production of the Higgs boson using the Optimal Observable method in the ditau decay channel with the ATLAS detector ATLAS arXiv:1602.04516

• Measurement of inclusive and differential cross

sections in the H → ZZ∗ → 4l decay channel at 13 TeV with the ATLAS detector ATLAS-CONF- 2017-032

4

Run 2 results 𝑔

Λ𝑅

VV Production (decay to 𝑔 ҧ 𝑔)

• Run 1: exclude spin 1 and 2
• Set limits on spin 0

anomalous couplings This analysis

Kinematics

• For a given 𝑛4ℓ, four-fermion system

in production or decay is defined by:

• 5 angles
• Two 𝑟𝑊𝑗

2 of difermion systems

• For the production+decay: two

𝐼𝑊𝑊 vertices

• 13 independent observables remain

𝐼 → 𝑎𝑎 → 4ℓ 𝑊𝐼 VBF

5

HVV amplitude

• 𝐵 𝐼𝑊𝑊 ~ 𝑏1

𝑊𝑊 + 𝑟𝑊1

2 +𝑟𝑊2 2

Λ1

𝑊𝑊 2

𝑛𝑊

1

2 𝜗𝑊

1

∗ 𝜗𝑊

2

∗ +

𝑏2

𝑊𝑊𝑔 𝜈𝜉 ∗ 1 𝑔∗ 2 ,𝜈𝜉 + 𝑏3 𝑊𝑊𝑔 𝜈𝜉 ∗ 1 ሚ

𝑔∗ 2 ,𝜈𝜉

• 𝑊𝑊 = 𝑎𝑎, 𝑋𝑋, 𝑎𝛿, 𝛿𝛿
• SM, tree level: 𝑏1

𝑎𝑎 = 𝑏1 𝑋𝑋 = 2, others = 0

• Assume 𝑏𝑗

𝑎𝑎 = 𝑏𝑗 𝑋𝑋, call it “𝑏𝑗”

• Assume no 𝑟2 cutoff for anomalous couplings
• Measure 𝑏2, 𝑏3, Λ1, Λ1

𝑎𝛿

• 𝑏2,3

𝑎𝛿,𝛿𝛿 are already constrained from onshell photons

• Parameterize as fractional cross section 𝑔

𝑏𝑗 = 𝑏𝑗 2𝜏𝑗 σ𝑘 𝑏𝑘

2𝜏𝑘

and relative phase 𝜚𝑏𝑗 = arg

𝑏𝑗 𝑏1

6

Tools

• JHUGen
• Generate samples with arbitrary couplings
• Τ

𝑕𝑕 𝑟ത 𝑟 → 𝑌 → Τ 𝑎𝑎 𝑋𝑋 → 4𝑔 for 𝑌 spin 0, 1, 2

• VBF, 𝑊𝐼, 𝑕𝑕𝐼 with 0, 1, or 2 QCD jets, 𝑢𝑢𝐼, 𝑐𝑐𝐼, 𝑢𝑟𝐼
• MELA—Matrix Element Likelihood Approach
• Matrix element calculations
• JHUGen for signal
• MCFM for background
• Calculate discriminants to distinguish hypotheses
• Reweight generated samples to different hypotheses

7

Contributions

• Background
• Τ

𝑟ത 𝑟 𝑕𝑕 → 𝑎𝑎

• 𝑎 + 𝑌
• Signal
• 𝑕𝑕𝐼, VBF, VH, 𝑢𝑢𝐼
• 𝐼𝑊𝑊 couplings in decay
• 𝐼𝑊𝑊 couplings in production and decay
• SM, anomalous, and interference contributions
• Want to isolate each component to constrain couplings
• 7 or 13 kinematic observables (+𝑛4ℓ for bkg separation)
• too many to use them all

8

Discriminants

• Two basic types of discriminants:
• 𝐸𝑏𝑚𝑢 =

𝑞𝑡𝑗𝑕 𝑞𝑡𝑗𝑕+𝑞𝑏𝑚𝑢

• Optimal to distinguish pure SM signal from alternate hypothesis
• Alternate hypothesis could be background, another coupling model, or

another signal production mode

• 𝐸𝑗𝑜𝑢 =

𝑞𝑗𝑜𝑢 𝑞𝑡𝑗𝑕+𝑞𝑏𝑚𝑢

• Together with 𝐸𝑏𝑚𝑢, optimal to also isolate the interference contribution
• 𝑞𝑡𝑗𝑕, 𝑞𝑏𝑚𝑢, 𝑞𝑗𝑜𝑢 are calculated through MELA using matrix

element probabilities

9

Discriminants 1

10

max 𝐸2𝑘𝑓𝑢

𝑎𝐼,𝑇𝑁, 𝐸2𝑘𝑓𝑢 𝑋𝐼,𝑇𝑁

• 𝐸2𝑘𝑓𝑢

Τ Τ 𝑊𝐶𝐺 𝑎𝐼 𝑋𝐼 = 𝑞

Τ 𝑊𝐶𝐺 𝑎𝐼∕𝑋𝐼

𝑞

Τ 𝑊𝐶𝐺 𝑎𝐼∕𝑋𝐼+𝑞𝐼𝑘𝑘

• Separate associated production from

QCD jets

• VBF-jet category:
• 𝐸2𝑘𝑓𝑢

𝑊𝐶𝐺,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑊𝐶𝐺,𝐶𝑇𝑁 > 0.5

• VH-jet category:
• 𝐸2𝑘𝑓𝑢

𝑎𝐼,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑎𝐼,𝐶𝑇𝑁 > 0.5

• r 𝐸2𝑘𝑓𝑢

𝑋𝐼,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑋𝐼,𝐶𝑇𝑁 > 0.5

• Untagged category:
• Everything else
• Use 𝐸2𝑘𝑓𝑢

𝑇𝑁 and 𝐸2𝑘𝑓𝑢 𝐶𝑇𝑁 to get optimal separation

for both extreme hypotheses

𝐸2𝑘𝑓𝑢

𝑊𝐶𝐺,𝑇𝑁

Discriminants 2

• Use 3D templates to

parameterize the signal and background for each category

• 𝑬𝒄𝒍𝒉, 𝐸𝑏𝑗, 𝐸𝑗𝑜𝑢
• 𝐸𝑐𝑙𝑕 =

𝑞𝑡𝑗𝑕 𝑞𝑡𝑗𝑕+𝑞𝑐𝑙𝑕

• Used for all 3 categories
• 𝑛4ℓ + decay kinematics

11

105 GeV < 𝑛4ℓ < 140 GeV

Discriminants 3

• 𝐸𝑐𝑙𝑕, 𝑬𝒃𝒋, 𝐸𝑗𝑜𝑢
• 𝐸𝑏𝑗 =

𝑞𝑡𝑗𝑕 𝑞𝑡𝑗𝑕+𝑞𝑏𝑗

• Tagged categories: use production×decay probabilities
• Untagged: use decay probabilities only
• Example: 𝐸0− for the 𝑔

𝑏3 analysis

12

Discriminants 4

• 𝐸𝑐𝑙𝑕, 𝐸𝑏𝑗, 𝑬𝒋𝒐𝒖
• 𝐸𝑗𝑜𝑢 =

𝑞𝑗𝑜𝑢 𝑞𝑡𝑗𝑕+𝑞𝑏𝑚𝑢

• Tagged categories: use production probabilities
• Untagged: use decay probabilities

13

Likelihood fit

• Assume real couplings, 𝜚𝑏𝑗 = 0 or 𝜌
• ggH, only one HVV vertex:
• 𝑞 𝑔

𝑏𝑗, Ω

~ 𝑏1𝐵1 + 𝑏𝑗𝐵𝑗 2 ∼ 𝑈0 Ω + 𝑏𝑗 𝑏1 cos 𝜚𝑏𝑗 𝑈

1 Ω + 𝑏𝑗

𝑏1

2

𝑈2 Ω

• VBF or VH, two HVV vertices
• 𝑞 𝑔

𝑏𝑗, Ω

~ (𝑏1𝐵1

𝑞𝑠𝑝𝑒 + 𝑏𝑗𝐵𝑗 𝑞𝑠𝑝𝑒)(𝑏1𝐵1 𝑒𝑓𝑑 + 𝑏𝑗𝐵𝑗 𝑒𝑓𝑑) 2

∼ ෍

𝑘=0 4

𝑏𝑗 𝑏1

𝑘

cos𝑘 𝜚𝑏𝑗 𝑈

𝑘 Ω

14

𝑔

𝑏𝑗 =

𝑏𝑗 2𝜏𝑗 𝑏1 2𝜏1 + 𝑏𝑗 2𝜏𝑗 + ⋯ 𝑏𝑗 𝑏1 = 𝑔

𝑏𝑗

𝑔

𝑏1

𝜏1 𝜏𝑗

Signal strength

• Want to decouple ratios of couplings 𝑔

𝑏𝑗 from the

signal strengths 𝜈𝑗

• Allow signal strength for production via fermion

couplings 𝜈𝑔 and boson couplings 𝜈𝑊 to float independently

• Constrained by category distribution of events

15

Event distribution

• First number is for SM, (second) is for 𝑔

𝑏3 = 1

• Use categorization for 𝑔

𝑏3 analysis, others are a bit different

• (In particular, fewer observed events in VBF-jets)

16

Results

• Scans for each

parameter

• 13 TeV only, and

combination with Run 1 result

17

More details: 𝑔

𝑏3 18

• 1D projections help to explain
• Small excess of events at smaller

values of 𝐸0−

𝑒𝑓𝑑

• Minimum away from 0
• 𝐸𝐷𝑄 has small excess on the right
• +0.3 is favored over -0.3
• Combine with Run 1:

minimum at 𝑔

𝑏3 = 0

More details: 𝑔

𝑏3

• VBF and VH are sensitive to

very small 𝑔

𝑏3

• Once 𝑔

𝑏3 ≳ 0.01, more

favorable to set 𝜈𝑊 → 0

19

𝜈𝑊

• Observe fewer

VBF and VH events than expected

• Best fit 𝜈𝑊 for

𝑔

𝑏𝑗 = 0 is < 1

(values on plots)

• Narrow minima

not as deep as expected

𝜈𝑊 = 0.76 𝜈𝑊 = 0.03 𝜈𝑊 = 0.20 𝜈𝑊 = 0.24

20

More details: 𝑔

𝑏3

• Animations: each frame uses

the best fit 𝜈𝑊 and 𝜈𝑔

• 𝑔

𝑏3 𝑊𝐶𝐺∕𝑊𝐼 are the cross section

fractions for those processes

• Watch what happens when

𝑔

𝑏3 ≲ 0.01

See ppt version for animations

• r separate indico attachment

21

Summary

• Constrain anomalous HVV couplings
• Decay information: 4 × more data

than in Run 1

• Production information: expected 1𝜏 exclusion of

small 𝑔

𝑏𝑗 values, observation falls a little short

• By the end of Run 2: expect production information

to give narrow 2𝜏 limits

22

Backup

23

Categorization

• 𝐸2𝑘𝑓𝑢

Τ Τ 𝑊𝐶𝐺 𝑎𝐼 𝑋𝐼 = 𝑞

Τ 𝑊𝐶𝐺 𝑎𝐼∕𝑋𝐼

𝑞

Τ 𝑊𝐶𝐺 𝑎𝐼∕𝑋𝐼+𝑞𝐼𝑘𝑘

• Separate associated production from QCD jets
• VBF-jet category:
• exactly 4 leptons
• 2 or 3 jets with at most one btag, or ≥ 4 jets with no btag
• 𝐸2𝑘𝑓𝑢

𝑊𝐶𝐺,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑊𝐶𝐺,𝐶𝑇𝑁 > 0.5

• VH-jet category:
• exactly 4 leptons
• 2 or 3 jets with at most one btag, or ≥ 4 jets with no btag
• 𝐸2𝑘𝑓𝑢

𝑎𝐼,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑎𝐼,𝐶𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑋𝐼,𝑇𝑁 > 0.5 or 𝐸2𝑘𝑓𝑢 𝑋𝐼,𝐶𝑇𝑁 > 0.5

• Untagged category:
• Everything else
• Use 𝐸2𝑘𝑓𝑢

𝑇𝑁 and 𝐸2𝑘𝑓𝑢 𝐶𝑇𝑁 to get optimal separation for both

extreme hypotheses

24

Discriminants table

25

Λ𝑅

26

• Λ𝑅 gives same kinematics, different mass shape
• Search from offshell region
• Limits assume Γ𝐼 = 4.1 MeV

CMS-HIG-14-036