observation standard model higgs and the cms lhc€¦ · l. dodd, u. wisconsin feb. 15 2018 1 ph.d....

Ph.D. DefenseL. Dodd, U. Wisconsin Feb. 15 2018 1

Laura Margaret Dodd

OBSERVATION OF STANDARD MODEL HIGGS BOSON DECAYS TO TAU LEPTONS AND A SEARCH FOR DARK MATTER WITH

THE CMS DETECTOR AT THE LHC

Ph.D. DefenseL. Dodd, U. Wisconsin Feb. 15 2018

About Me • Laura Margaret Dodd

• B.S. in physics from Duke University

• Worked in ATLAS group

• Entered UW-Madison physics Ph.D. program in 2012

• Advisor: Wesley Smith

• Stationed at CERN between 2014 and 2017

2


Roadmap

• Motivation and theory

• Experiment and background

• Observation of standard model (SM) Higgs boson to tau pairs

• Search for dark matter (DM) pairs in association with SM Higgs boson in tau pair final state.

• Summary

3


Standard Model (SM)

4

• Fermions (spin-1/2)

• 3 generations of matter

• 6 quarks, and 6 leptons

• 3 charged and 3 neutral leptons

• In SM, neutrinos are massless, unknown extension needed

• Bosons (integer spin) - force mediators

• gluons -> strong force

• γ -> EM force

• W±/Z0 -> Weak force

massless

massless

massive } unified electroweakforce in SM

W±/Z0 sometimes denoted V


Standard Model (SM): Higgs I

5

• Higgs Mechanism

• Scalar doublet is added

• Gauge bosons W±/Z0 acquire mass through spontaneous symmetry breaking and nonzero vacuum expectation

• Massive scalar boson predicted: Higgs boson

• Higgs couples to fermions and therefore fermions gain mass

• 2 main production mechanisms considered:

• gluon fusion (ggH/ggF) and vector boson fusion (VBF)

gluon fusion (ggH)

vector boson fusion (vbf)rate is factor of ten smaller than ggH


Standard Model (SM): Higgs III

6

• In 2012 CMS and ATLAS measured the cross sectional rate of the Higgs coupling to tau leptons and the Higgs coupling to b quarks.

• Signal strength (μ) is defined as observed rate / expected SM rate.

• CMS measured the rate of H→ττ to be μ= 0.88±0.30 times the SM expectation.

• ATLAS measured the rate of H→ττ to be μ=1.41 ±0.40 times the SM expectation

arXiv:1401.6527 [hep-ex]

ATLAS+CMS measure H(bb) μ=0.70 ± 0.29


Dark Matter

7

• Dark matter (DM) exists; the majority of matter in the universe is DM.

• The fundamental nature of DM is not known

• Weakly interacting massive particle (WIMP) may interact with SM through the Higgs sector, as in Higgs-portal models. WIMP DM is denoted χ.

• Ι focus on a collider “mono-Higgs” signature e.g. Higgs+ nothing detected

• Two models examined provide a non-resonant and a resonant handle

2HDM-Z’

baryonic Z’

non-resonant

resonant


τ lepton decays

8

⌫⌧

e µ d s

W�

⌧�

⌫e ⌫µ u u

1

• Tau lepton heaviest lepton in the SM: 1.77 GeV • Most strongly coupled lepton to Higgs

• Tau lifetime is ~3×10-13 seconds • Taus can decay to hadrons (τh) as well as e,μ

π+ 139 MeV π0 135 MeV

L A R G E H A D R O N C O L L I D E R

OFF ICE

DETECTOR

9


LHC• 27 km proton-proton or

heavy-ion collider

• In 2016, operated at 13 TeV center of mass energy

• CMS and ATLAS, general purpose detectors

• ALICE (heavy ion) and LHCb (b-physics)

• Protons accelerated in stages

• Injected into LHC at 450 GeV and accelerated to 6.5 TeV by 400 MHz RF cavities

10


Cross section and Luminosity

11

0.1 1 1010

-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

102

103

104

105

106

107

108

109

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

101

102

103

104

105

106

107

108

109

σσσσZZ

σσσσWW

σσσσWH

σσσσVBF

MH=125 GeV

HE

LHC

WJS2012

σσσσjet

(ET

jet > 100 GeV)

σσσσjet

(ET

jet > √√√√s/20)

σσσσggH

LHCTevatron

eve

nts

/ s

ec f

or L

= 1

03

3 c

m-2s

-1

σσσσb

σσσσtot

proton - (anti)proton cross sections

σσσσW

σσσσZ

σσσσt

σ

σ

σ

σ

(( ((nb

)) ))

√√√√s (TeV)

{

1 ASU1 0

Dy1 Ju

n1 Ju

O1 AuJ

1 6eS1 2

ct1 1

ov1 D

ec

DDte (87C)

0

10

20

30

40

50

60

7otD

O ,n

teJ

UDte

d L

um

LnoVLt

y (fb−1)

× 50

DDtD included fUom 2010-03-30 11:22 to 2017-08-08 19:00 87C

2010, 7 7eV, 45.0 pb−1

2011, 7 7eV, 6.1 fb−1

2012, 8 7eV, 23.3 fb−1

2015, 13 7eV, 4.2 fb−1

2016, 13 7eV, 40.8 fb−1

2017, 13 7eV, 14.3 fb−1

0

10

20

30

40

50

60

CMS ,ntegrated LumLnosLty, SS

To increase luminosity: decrease the collision angle (β),

maximize the number of particles per bunch (Νb), and

squeeze each bunch (~𝛜) L =R

�(1)

2016 pp: ~27 interactions/ bunch crossing

C O M PA C T M U O N S O L E N O I D ( C M S )

12


CMS Overview

13

3.8 T 19.5 kA superconducting

solenoid cooled to 4.7K provides magnetic field to

measure momenta of particles

pTΦ

+z-z

η

θ

p

Hermetic, calorimeters

inside solenoid


Tracker

14

• Silicon tracking system is the first layer within CMS

• Three pixel barrel layers at 4 cm, 7 cm and 10 cm, and two endcap layers extend to |η|<1.5.

• The silicon strips outside the pixel extend coverage up to |η|<2.5


Electromagnetic Calorimeter

15

• Lead tungstate (PbWO4) crystals

• Measures scintillation light

• Extends up to |η| =3

• 26 radiation lengths

• Overall resolution

• Electrons: 0.4% (0.8%) in the barrel (endcaps)

• Photons: roughly ~2%(~4%) in the barrel (endcaps)

density 8.28 g cm-3

radiation length 0.89 cm

molière radius 2.2 cm

interaction length 20.27 cm

light yield ~100 γ /MeV⇣ �

E

⌘2=

✓2.8pE

◆+

0.122

E+ 0.0032 (1)


Hadronic Calorimeter

16

• Extends to |η|<5.2

• Non-ferromagnetic brass absorbers and plastic scintillators for HB/HE

• 16.42 cm nuclear interaction length, and 1.49 cm radiation length

• HB is 12 interaction lengths deep

• HF is steel and quartz fiber Cherenkov-based, 10 interaction lengths deep

HB/HE

HF


Muon System

17

• 1.9 T magnetic field

• Three gaseous chamber technologies until |η|<2.4

• drift tubes (DTs), cathode strip chambers (CSCs), resistive plate chambers (RPCs).

• 10% resolution muons |η|<2.4 and pT<200 GeV muon system only

• Including the tracker measurement improves resolution to less than 2%

|η|<2.4 and pT>200 GeV muon system resolution still ~10%

global fit closer to 5%


Muon Track-Finder Layer

MPC

CSC DT

LB

RPC

Mezz

Endcap

CuOF

Barrel

Global Muon Trigger

Global Trigger

TwinMux

CPPF

Overlap

ECAL HCAL HB/HE

HCALHF

OSLB

Calo Trigger Layer 1

Calo Trigger Layer 2

Calorimeter Trigger Muon Trigger

CMS Trigger System

18

HLT

output rate 100kHz

Save to Diskoutput rate

1kHz

Input beam crossing rate 40MHz

single muon trigger pt>22 or pt>24

single muon + tau

pt(mu) >18 and pt(tau)>20 (tau)>20

single electron pt >25

double hadronic

tauboth pt>35

• Two level trigger system: hardware-based Level-1 and software-based High Level Trigger (HLT)


Particle Flow

19

• Particle Flow (PF) algorithm, used by CMS, links the subsystems together and creates four-vectors for reconstructed physics objects.

• From tracker hits, muon hits, and calorimeter energy deposits, physics “object” four-vectors are created

primary vertex: the collision(vertex) from which our objects are

linked.

PF jets: composed mostly of

charged hadrons, photons, and neutral

hadrons.

pT Transverse Momentum


Electron/Muon Identification

20

• Muons leave tracks in both the tracker and muon system, with no linked calorimeter deposit

• Muon is identified in both the tracker and the muon system

• Muons cut based identification >90% efficient

• Electrons have a track+ ECAL deposit, with no HCAL/Muon deposits

• MVA based identification • signal electrons: 80% efficiency

signal isolation cone charged, neutral ΔR<0.3 (0.4) for electrons (muons)

pileup subtraction term


Hadronic Tau Identification

21

• 55% of tau lepton decays are hadronic τh

• Working point efficiencies and mis-identification probabilities (probability for a jet to fake a tau given loose jet identification requirements)

Eff (%) Mis-idprob (%)

tight 55 8x10-3

med 60 1x10-2

loose 65 2x10-2

anti-e 80 10-3

anti-mu 90 10-3


Hadronic Tau Reconstruction

22

• Hadronic taus are seeded from PF Jets with ΔR=0.4

• Hadron plus strips (HPS) algorithm used to reconstruct hadronic taus

• Dynamic strip reconstruction added in 2015, strip size varies

π0

π+/-π+/- π+/-

π+/-

1-prong 1-prong 1-pi0

3-prongDECAYMODE


MET and Jet Identification

23

• PF Jets use the Anti-kt algorithm (AK) that clusters the harder particles first, resulting in more circular jets. Default cone size is ΔR=0.4 . They are further corrected with Jet Energy Corrections (JEC).

• Type-1 missing transverse energy (MET) is the negative sum of the transverse momenta of recalibrated jets, and PF objects (e.g. charged hadrons, electrons, and muons).

• Jets are tagged as a b quark jet by identifying possible secondary vertices


Simulation and Perturbation

24

PDF NNPDF 3.0 and NNPDF3.1

Hard scatter POWHEG, MadGraph5+aMC@NLO

Parton Shower Pythia

Hadronization Pythia

Decay Pythia + GEANT

Detector Simulation GEANT

Perturbative expansion of a coupling constant: α<<1

example: perturbative QCD

NLO

1 + α + α2 + α3

Leading order (LO)

Next toLO

(NLO)

Next to NLO

(NNLO)


Standard Model H→τ τ

25


Analysis Overview

26

• Goals: Measure coupling strength of Higgs boson to tau leptons.

• Target gluon fusion and vector boson fusion productions to measure rate

• Four channels τeτh (denoted eτh) , τμτh (denoted μτh), τhτh, and eμ

• I focus on eτh, μτh, together denoted as ℓτh

• Combined ℓτh make up over 50% of analysis sensitivity

• Two-dimensional signal extraction is used:

• One dimension: invariant mass distribution

• Second dimension: varies on targeted production

gluon fusion (ggH) 13 TeV cross section: 48.58 pb

vector boson fusion (vbf)13 TeV cross section: 3.78 pb


ττ Backgrounds

27

Drell-Yan Z→ ℓℓ

5700 pbtt → WbWb

830 pb

and inclusive cross section

fake τh

W→ℓν 61000 pb

genuine tau background

mis-reconstructed taubackground

resonantnon-resonant

QCD/multijet~ 1 mb

ℓ=e,μ,τ


Kinematic Selections

28

• Higgs (ττ) final decay products mostly low momentum

• τ→μ and τ→e decay, majority of the τ momentum to go to neutrinos. Expect falling distribution from 15 GeV.

• τ→τh expect falling distribution from 30 GeV.

• Kinematic selections chosen to increase acceptance

• However to reduce fake background, choose pT(τh)>30 in all channels

channel pT

eτhe>26τh>30

μτhμ>20τh>30

τhτhτh>50τh>40

eμ e>13(24) μ>24(15)

η within tracker range excluding eμ

full selection table in thesis

thisthesis


Mass Reconstruction (mττ)

29

• The di-tau mass (mττ) is reconstructed using a kinematic fit in most categories

• Allows for some separation of Z→ττ and H→ττ, and shifts the H→ττ peak to 125 GeV.

• The kinematic fit takes as input: MET, MET uncertainty, each tau candidate’s four vector. It assumes all MET in the event is from neutrinos in tau decay.

given all MET in event comes from tau decay, kinematic fit finds most likely original mass of di-tau system and is

denoted mττ. Visible mass is mvis.


Background Methods I

30

Z->ττ simulation:Madgraph Drell-Yan

Samples.Corrections derived

from highly pure Z →μμ data

sample.

DY ℓ->τ fakes simulation:Madgraph Drell-Yan

Samples.Corrections derived from highly pure Z →μμ data sample. Additional ℓ->τ fake corrections applied.

W+Jets/diboson:Normalization controlled by

data agreement in sideband for

ℓτQCD:

derived from data in sideband region, for ℓτ,

ττ, eμ

TTbar:NLO MC used. Normalization

controlled by data agreement in

sideband for all channels

prefit


Background Methods IIℓτ: QCD estimate is taken from same-sign region and

scaled by factor to go to opposite-sign region

sig. reg.

ℓτ: W+Jets estimate is taken from high MT,1

w+jetcontrol region

MT,1

MT,1 =q2p`TE

missT (1� cos(�)) (1)

<latexit sha1_base64="yzBNdlG8hT3u0AfAT5R3ThXBOeY=">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</latexit><latexit sha1_base64="yzBNdlG8hT3u0AfAT5R3ThXBOeY=">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</latexit><latexit sha1_base64="yzBNdlG8hT3u0AfAT5R3ThXBOeY=">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</latexit>

we expect no signal in the same-sign region


Event Selection I

32

MT,1

• Vetoing extra leptons, and requiring well-isolated objects results in the largest decrease in events.

• Next the MT,1<50 GeV cut is applied

Looking for 109.9 events out of 492668 observed

events


Event Selection II

33

• Expect roughly 0.2% of collected events are from gluon fusion and vector boson fusion: 1650 expected signal with 664000 expected background

• Need to increase sensitivity further, define 3 categories: 0jet (targets ggH), boosted (targets ggH), vbf (targets VBF).

0jet no jets with pT>30 GeV

vbf

≥2 jets with pT>30 GeVmjj>300 GeV

pT(ττ)>50 GeV pT(μ)>40 GeV (μτh only)

boosted >0 jets andnot VBF


Observed Events

34

Categorization helps separate ggH and VBF signal. Look at mττ and mvis distributions for shape discrimination

Looking for 29.5 events out of

2927 observed events


Hττ 2D Categories

35

τhτh μτh eτh eμ0jet mττ mvis:τ DM mvis:τ DM mvis : μ pT

boosted mττ : H pT mττ : H pT mττ : H pT mττ : H pT

vbf mττ : mjj mττ : mjj mττ : mjj mττ : mjj

The signal/background increases as a function of total higgs pT (H pT)

The signal/background increases as a

function of leading di-jet invariant mass mjj

has better separation of fakes

in visible mass (mvis) and tau decay mode

(τ DM): next slide

p⌧⌧T = |~pLT + ~pL0

T + ~EmissT |

and 12addl’

sidebands for norm.of bkgs.


visible mass: tau decay mode

36

0jet: better Z →μμ separation in visible mass

than a kinematic fit

visible mass [GeV]40 60 80 100 120 140 160 180 200

Obs

/Exp

0.60.8

1

1.2

1.4

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

Even

ts

0

5000

10000

15000

20000

25000Observed

ττ→Zµµ→Z

ElectroweakQCDtt

ττ→10xSM H(125)

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

visible mass [GeV]40 60 80 100 120 140 160 180 200

Obs

/Exp

0.60.8

1

1.2

1.4

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

Even

ts

0

5000

10000

15000

20000

25000Observed

ττ→Zµµ→Z

ElectroweakQCDtt

ττ→10xSM H(125)

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

SVfit mass [GeV]40 60 80 100 120 140 160 180 200

Obs

/Exp

0.60.8

1

1.2

1.4

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

Even

ts

0

2000

4000

6000

8000

10000

12000

14000

16000

18000 Observedττ→Zµµ→Z

ElectroweakQCDtt

ττ→10xSM H(125)

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

SVfit mass [GeV]40 60 80 100 120 140 160 180 200

Obs

/Exp

0.60.8

1

1.2

1.4

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

Even

ts

0

2000

4000

6000

8000

10000

12000

14000

16000

18000 Observedττ→Zµµ→Z

ElectroweakQCDtt

ττ→10xSM H(125)

(13 TeV)-1 2016, 35.9 fbhτµτ

CMSPreliminary

ττ μτ eτ eμ

0jet mττ mvis:τ DM mvis:τ DM mττ : μ pT

very few μ->τ fakes in 3 prong, higher S/B

use tau decay mode

mττ

mvis


2D Distribution

37

“0jet eτ visible mass versus tau decay mode”

1-prong tau visible mass distribution

1-prong-1pi0 tau visible mass distribution

3-prong tau visible mass distribution

mass variable used for

extractionGeV range for mass variablein a bin


Channel and Category • The SM Higgs excess is seen when sorted by significance of each bin

• VBF is the most sensitive category; μτh eτh are second and third significant channel.

38

channellog(S/(S+B))

3− 2.5− 2− 1.5− 1− 0.5− 0

Even

ts

1−10

1

10

210

310

410

510

610

710

VBF Boosted

0 jet ττ→H

Uncertainty Observed

(13 TeV)-12016, 35.9 fb

CMSPreliminary

log(S/(S+B))1.5− 1− 0.5−

00.20.40.60.8

11.21.41.6 (Data - bkg)/bkg

)/bkgττ→(HBkg. unc.

channel category


H(ττ) weighted mass plot • The mass distributions weighted by S/(S+B) show a

visible SM Higgs peak

39


Energy Scale Uncertainties

40

ν is not included in TES, MET ES

needed.

π- are included in

1.2% Tau ES

νν

τ+

τ-H(125)

π+π+ π-

μ-

ν

μ- have negligible ES

compared with MET, Jet and

Tau ES

Jets in the event affect the MET

measurement. JET ES added to MET

calculation.

Energy Scales (ES) Uncertainty Example: a 50 GeV tau could be +/- 0.4 GeV

jet


Rate measurement

41

• Measured versus expected rates are shown by category and channel

• Measured SM Higgs(ττ) rate is 1.09 +0.27-0.26 of the expected SM rate.

• When shapes are included in the statistic, the observed (expected) significance of the measured H(ττ) decay is 4.9 (4.7) sigma compared to SM expectation without Η(ττ) decay


fermionic-vector coupling (κ)• κ, is the ratio of the measured

coupling to the SM expectation.

• The categories separate ggH and VBF for signal production measurements. The measured rate can be broken down as a function of the VBF and ggH process. The vector coupling rate, κV, is compared to the fermionic coupling rate, κf.

• Explicitly, we measure ggH rate at 1.07+0.45-0.41 times the SM expectation, and the VBF rate at 1.06 +0.42-0.41 times the SM expectation.

42


[pb]

σPr

oduc

tion

Cro

ss S

ectio

n,

4−10

3−10

2−10

1−10

1

10

210

310

410

510

CMS PreliminaryJuly 2017

All results at: http://cern.ch/go/pNj7W

n jet(s)≥

Z

n jet(s)≥

γW γZ WW WZ ZZµll, l=e,→, Zνl→EW: W

qqWEW qqZ

EWWW→γγ

γqqWEW

ssWW EW

γqqZEW

qqZZEW γWV γγZ γγW tt

=n jet(s)

t-cht tW s-cht γtt tZq ttW ttZ ttttσΔ in exp. HσΔTh.

ggH qqHVBF VH ttH HH

CMS 95%CL limits at 7, 8 and 13 TeV

)-1 5.0 fb≤7 TeV CMS measurement (L )-1 19.6 fb≤8 TeV CMS measurement (L )-1 35.9 fb≤13 TeV CMS measurement (L

Theory prediction

Cross section summary

43

[pb]

σPr

oduc

tion

Cro

ss S

ectio

n,

4−10

3−10

2−10

1−10

1

10

210

310

410

510



n jet(s)≥

Z

n jet(s)≥


qqWEW qqZ

EWWW→γγ

γqqWEW

ssWW EW

γqqZEW


=n jet(s)





Theory prediction

[pb]

σPr

oduc

tion

Cro

ss S

ectio

n,

4−10

3−10

2−10

1−10

1

10

210

310

410

510



n jet(s)≥

Z

n jet(s)≥


qqWEW qqZ

EWWW→γγ

γqqWEW

ssWW EW

γqqZEW


=n jet(s)





Theory prediction

VBF measured uncertainty

added to CMS cross section summary plot

H→ττ measured in very good agreement with the SMat a high level of precision! C

ross

Sec

tion σ

(pb)

previous slide:VBF rate at 1.06 +0.42-0.41 times

SM expectation


Combination with 8 TeV

44

• Conservative combination with the 8 TeV CMS H(ττ) analysis is performed. The observed and expected significance of the H(ττ) decay increases to 5.9 (5.9) sigma as compared with no SM H(ττ).


Mono-Higgs(ττ)

46


Mono-Higgs Overview

47

• Produce a limit on the number of Higgs bosons (ττ) that decay in association with dark matter.

• use eτh, μτh, τhτh, final states onlysimilar background methods as SM Ηττ3 signal regions (eτh, μτh, τhτh) and 5 control regions for background normalization

• Higgs and MET are roughly back to back, instead of roughly collinear in SM Higgs. The previously used kinematic fit is unusable. Di-boson backgrounds play larger role.


Mono-Higgs Overview

48

events with b-tagged

jets (60% efficiency WP)

are vetoed


μτh 2HDM

Observable: MT,tot

49

• Total transverse mass (MT,tot) used for extraction, signal distributions beyond 260 GeV, background mostly below

(13 TeV)-1 2016, 35.9 fbhτhτ

CMSPreliminary

Total [GeV]TM0 20 40 60 80 100 120 140 160 180 200

Even

ts0

50

100

150

200

250Observed

ττ→Z other→Z

WJetsVV Single TopggH, vbfH, ZHQCDtt

+Jetsνν→Zbkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτhτ

L.DoddPh.D. ThesisUW-Madison

τhτh

MT,tot [GeV]


Limit Extraction

50 (GeV)T,totM

0 200 400 600 800 1000 1200 1400 1600 1800 2000

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτeτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310

410 Observedττ→Z

ll→ZW+jetsMultiboson/Single TopQCD/multijetSM 125 GeV Higgstt

bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτeτ

L. DoddPh.D. ThesisUW-Madison

(GeV)T,totM0 200 400 600 800 1000 1200 1400 1600 1800 2000

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτμτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310



bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτμτ


(GeV)T,totM0 200 400 600 800 1000 1200 1400 1600 1800 2000

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτhτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310

Observedττ→Z

+jetsνν→ZW+jetsMultiboson/Single TopQCD/multijetSM 125 GeV Higgstt

bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτhτ


• Systematics roughly identical to SM Higgs to ττ

• Limited number of simulated events

• Background precision decreases in tails of distribution

• W+jets, WW, and ZZ higher-order correction uncertainties applied, dependent on generated boson pT


Event Yield

51

background yields in sensitive regiondominated by statistical uncertainties

Can expect up to 5 events in each categoryfor expected Mono-Higgs signal


Z’-2HDM Model Exclusion

52

[GeV]Z'M600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900

[GeV

]A

M

300

350

400

450

500

550

600

650

1

10

210

310Observed Limit (95% CL)Expected Limit (95% CL)

σExpected Limit +/- 1

(13 TeV)-135.9 fbCMS Preliminary

exclude MZ’ ={700 -1200 GeV} and MA={300-375 GeV}

Mono(ττ) Mono(ττ)+Mono(γγ)

combination of γγ+ττ increases exclusion:

MZ’ ={600 -1350 GeV} and MA={300-400 GeV}

No significant excess compared to

SM expectation


(GeV)Z`M10 210 310 410

) (fb

)χχ

H→ B

Z'

→(p

pσ

95%

C.L

.

2−10

1−10

1

10

210 Observed Limitττ+γγ Expected Limitττ+γγ 1 std. dev.ττ+γγ

Observed Limitττ

Observed Limitγγ

Baryonic Z’ Model Exclusion

53

combination of γγ+ττ increases

exclusion range

No significant excess compared to

SM expectation

currentlypreparing for publication

Mono(ττ)+Mono(γγ)


Summary

54

• SM H(ττ) measurement sees Η(ττ) decay at high-level of precision consistent with prediction

• No significant excess compared to the standard model expectation in dark matter mono-Higgs search


END

55


Backup

56

A thesis is never finished, only abandoned.—Marià Cepeda,

quoting Juan Alcaraz, quoting someone else


SMH backup

57


P-value and Significance• 125 GeV SM Higgs

• Expected (postfit) significance is 4.7 sigma

• Observed significance is 4.9 sigma

• Obs. p-value ~0.00000006

58


Visible Tau Energy Scale • Visible Tau Energy Scale

• Includes track and ECAL measurement • Assigned 1.2% Up/Down uncertainty

• TES constrained to small values (0.3%) • Correlate the taus whatever eta/pt/

category • Core of Zττ has taus with pT<70 GeV

highest boost still has mostly low pT taus

• CMS measures tracks well, more final states included than 1.2% measurement.

59

tau decay mode correction

1 prong -1.8%

1 prong 1 pi0 +1%

3 prong +0.4%


Systematic Uncertainties

60

• Full list included in thesis

• Systematics affecting normalization

• Luminosity measurement

• Trigger efficiencies

• Hadronic tau reconstruction

• Systematics affecting shapes

• Energy Scales (discussed next)

• Drell Yan reweighting to match observed distributions

• W+jet and QCD variation

• Higgs momentum distributions (theory)

• Limited number of events (bin-by-bin)


(GeV)vism60 80 100 120

Prob

abilit

y de

nsity

0

0.05

0.1

0.15

0.2

0.25

0.3

13 TeVCMS Simulation Preliminary

)hτµ (0jet, µµ→Z

1 pr

ong

0π

1 pr

ong

+ 3

pron

gs

(GeV)vism60 80 100 120

Prob

abilit

y de

nsity

0.01

0.02

0.03

0.04

0.05

0.06

0.0713 TeVCMS Simulation Preliminary

)hτµ (0jet, ττ→ggH

1 pr

ong

0π

1 pr

ong

+ 3

pron

gs

(GeV)ττm50 100 150 200

(GeV

)jj

m

400

600

800

1000

1200

1400

1600

1800

Prob

abilit

y de

nsity

0

0.01

0.02

0.03

0.04

0.05


)hτµ (VBF, ττ→qqH

(GeV)ττm50 100 150 200

(GeV

)jj

m

400

600

800

1000

1200

1400

1600

1800

Prob

abilit

y de

nsity

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16


)hτµ (VBF, ττ→Z

(GeV)ττm50 100 150 200

(GeV

)ττ Tp

0

50

100

150

200

250

300

350

400

Prob

abilit

y de

nsity

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

0.018

0.02

0.022


)hτµW+jets (boosted,

(GeV)ττm50 100 150 200

(GeV

)ττ Tp

0

50

100

150

200

250

300

350

400

Prob

abilit

y de

nsity

0

0.005

0.01

0.015

0.02

0.025

0.03


)hτµ (boosted, ττ→ggH

2-Dimensional Distributions

61

0jet boosted vbf

sign

alba

ckgr

ound

sign

alba

ckgr

ound

Ph.D. DefenseL. Dodd, U. Wisconsin Feb. 15 2018 (GeV)vismU

nder

flow

60-6

565

-70

70-7

575

-80

80-8

585

-90

90-9

595

-100

100-

105

105-

110

Ove

rflow

Und

erflo

w60

-65

65-7

070

-75

75-8

080

-85

85-9

090

-95

95-1

0010

0-10

510

5-11

0O

verfl

owU

nder

flow

60-6

565

-70

70-7

575

-80

80-8

585

-90

90-9

595

-100

100-

105

105-

110

Ove

rflow

Exp.

(Obs

.-Exp

.)

1−01

Even

ts/b

in

1

10

210

310

4101 prong 0π1 prong + 3 prongs

Observed µτhτ→Z DY others +jetstt DibosonW+jets QCD multijet 125ττ→H Uncertainty

(13 TeV)-135.9 fbCMS Preliminary , 0 jethτµ

(GeV)vismUnd

erflo

w60

-65

65-7

070

-75

75-8

080

-85

85-9

090

-95

95-1

0010

0-10

510

5-11

0O

verfl

owU

nder

flow

60-6

565

-70

70-7

575

-80

80-8

585

-90

90-9

595

-100

100-

105

105-

110

Ove

rflow

Und

erflo

w60

-65

65-7

070

-75

75-8

080

-85

85-9

090

-95

95-1

0010

0-10

510

5-11

0O

verfl

ow

Exp.

(Obs

.-Exp

.)

1−01

Even

ts/b

in1

10

210

310

410 1 prong 0π1 prong + 3 prongs

Observed eτhτ→Z DY others +jetstt DibosonW+jets QCD multijet 125ττ→H Uncertainty

(13 TeV)-135.9 fbCMS Preliminary , 0 jethτe

0jet distribution: eτ μτ

62

eτ 0jet: mvis : τ decay mode

μτ 0jet:mvis : τ decay mode

Ph.D. DefenseL. Dodd, U. Wisconsin Feb. 15 2018 (GeV)ττmU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-13

013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

ow

Exp.

(Obs

.-Exp

.)

1−012

Even

ts/b

in

2−10

1−10

1

10

210

310

410 > 0 GeVττT

p > 100 GeVττT

p > 150 GeVττT

p > 200 GeVττT

p > 250 GeVττT

p > 300 GeVττT

p


(13 TeV)-135.9 fbCMS Preliminary , Boostedhτµ

(GeV)ττmUnd

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Exp.

(Obs

.-Exp

.)

1−01

Even

ts/b

in2−10

1−10

1

10

210

310

410 > 0 GeVττT

p > 100 GeVττT

p > 150 GeVττT

p > 200 GeVττT

p > 250 GeVττT

p > 300 GeVττT

p


(13 TeV)-135.9 fbCMS Preliminary , Boostedhτe

boosted distribution: eτ μτ

63

eτ boosted: mττ : H pt

μτ boosted:mττ : H pT


nder

flow

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Exp.

(Obs

.-Exp

.)

01234

Even

ts/b

in

2−10

1−10

1

10

210

310 > 300 GeVjjm > 700 GeVjjm > 1100 GeVjjm > 1500 GeVjjm


(13 TeV)-135.9 fbCMS Preliminary , VBFhτµ

(GeV)ττmUnd

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Exp.

(Obs

.-Exp

.)

1−01

Even

ts/b

in2−10

1−10

1

10

210



(13 TeV)-135.9 fbCMS Preliminary , VBFhτe

vbf distribution: eτ μτ

64

eτ vbf: mττ : di-jet mass

μτ vbf:mττ : di-jet mass


(GeV)Hm110 115 120 125 130 135 140

SMσ/

σ95

% C

L lim

it on

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8Observed Expected

Expectedσ1± Expectedσ2±

Observed Expected

Expectedσ1± Expectedσ2±CMS

)-1Preliminary(35.9 fb

Combined Limit information• We see significant excess

centered near 125 GeV combined asymptotic limits

65

(GeV)Hm110 115 120 125 130 135 140

SMσ/

σ95

% C

L lim

it on

0

0.5

1

1.5

2

2.5

limits_cmb.json:exp0 limits_em.json:exp0

limits_et.json:exp0 limits_mt.json:exp0

limits_tt.json:exp0

limits_cmb.json:exp0 limits_em.json:exp0

limits_et.json:exp0 limits_mt.json:exp0

limits_tt.json:exp0

CMS)-1Preliminary(35.9 fb

• ττ most sensitive channel, followed by μτ


W+Jets CR: ℓτ postfit

66

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

2

4

6

8

10

12310×

Observed eτhτ→Z

DY others +jetsttDiboson W+jets

QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

1

2

3

4

5

6

7

8

9310×

Observed eτhτ→Z


QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Obs

./Exp

.0.8

1

1.2Ev

ents

/bin

02468

10121416182022

310×

Observed µτhτ→Z


QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

2

4

6

8

10

12

14

16

310×

Observed µτhτ→Z


QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

0jet 0jetboosted boosted

eτ μτeτ μτ80 GeV <MT < 200 GeV

signal region is MT<50 GeVMT =q2pTEmiss

T (1� cos(��))


(GeV)ττm50 100 150 200

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

310×

Observed

µτhτ→Z

DY others

+jetsttDiboson

W+jets

QCD multijet

Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

(GeV)vism50 100 150 200

Obs

./Exp

.0.8

1

1.2Ev

ents

/bin

0

2

4

6

8

10

12

14

16310×

Observed

µτhτ→Z

DY others

+jetsttDiboson

W+jets

QCD multijet

Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

(GeV)ττm50 100 150 200

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

100

200

300

400

500

600

700Observed

eτhτ→Z

DY others

+jetsttDiboson

W+jets

QCD multijet

Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

(GeV)vism50 100 150 200

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

310×

Observed

eτhτ→Z

DY others

+jetsttDiboson

W+jets

QCD multijet

Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

QCD CR: ℓτ postfit

67

0jet 0jetboosted boosted

eτ μτeτ μτAnti-Isolated ℓ Region


S/(S+B) Weighted M_VIS Plot

68

(GeV)vism0 20 40 60 80 100 120 140 160 180 200

S/(S

+B) w

eigh

ted

even

ts /

GeV

5

10

15

20

25

30

35

40

ττ→Z

W+jets

QCD multijet

Other bkg.

ττ→H

Uncertainty

Observed

(13 TeV)-12016, 35.9 fb

CMSPreliminary

µ, ehτµ, hτ0 jet: e

(GeV)vism0 50 100 150 200

0.4−

0.3−

0.2−

0.1−

00.10.20.30.4

Data - bkg

ττ→H

Bkg. unc.


Visible TES treatment

69

tau decay mode correction

1 prong -1.8%

1 prong 1 pi0 +1%

3 prong +0.4%

core of Zττ has taus with pT<70 GeV

highest boost still has mostly low

pT taus

0-jet Boosted: ptH > 300 GeV

τ pt τ pt

τ η

τ η


Tau eta vs Tau pt per category

70

Boosted:200<ptH<250GeV

Boosted:100<ptH<150GeV Boosted:150<ptH<200GeV

Boosted:250<ptH<300GeV Boosted:ptH>300GeV

0jet


0-jet τ Decay Mode 2D

71

• In 0-jet category for ℓτ channels, ZL distribution is mostly confined to 1-prong decay mode and 1-prong 1-pi0. 3 prong is devoid of ZL background.

• Tau ℓ->τ fake scale factors were measured inclusively in pT and hard to extrapolate differentially

• Tau Energy Scale (TES) is better measured when versus by decay mode

(GeV)vism60 80 100 120

Prob

abilit

y de

nsity

0.01

0.02

0.03

0.04

0.05

0.06


)hτµ (0jet, ττ→ggH

1 pr

ong

0π

1 pr

ong

+ 3

pron

gs


Zee/Zμμ Treatment • The e->τ, μ->τ fakes are further corrected in energy scale (ES) and yield

in eτ and μτ. Versus by decay mode in 0-jet necessitated changes.

• The 1.2% τh ES is not applied to ℓ->τ fakes, the ES measured for ℓ->τ fakes is shown in the table below. Additionally τh anti-lepton discriminator scale factors were measured inclusively, not split by decay mode. Scale factors per decay mode are applied.

72

1-Prong Decay Mode 1-Prong+pi0 3-Prong

μτ τ pt ES -0.2% +/- 1.5% 1.5% +/- 1.5% 0 +/- 1.5%

μ->τ fake rate correction

0.75 (25%) 1.0 (25%) -

eτ τ pt ES 0 +/- 3% 9.5% +/- 3% 0 +/- 3%

e->τ fake rate correction

0.98(12%) 1.2(12%) -


Full Kinematic Selections

73


Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

20

40

60

80

100

310×

Observed µτeτ→Z

DY others +jetstt

Diboson W+jets

QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Other CR: ttbar, ττ QCD postfit

74

• OS, anti-isolated τ, svMass

• 0jet (mττ: 0-300 GeV)

• boosted/vbf (mττ: 0-250 GeV)

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

0.2

0.4

0.6

0.8

1

1.2

310×

Observed hτhτ→Z

DY others +jetstt

Diboson W+jets

QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Obs

./Exp

.

0.8

1

1.2Ev

ents

/bin

0

5

10

15

20

25

30

35

40310×

Observed hτhτ→Z

DY others +jetstt

Diboson W+jets

QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

Obs

./Exp

.

0.8

1

1.2

Even

ts/b

in

0

10

20

30

40

50

60

70

310×

Observed hτhτ→Z

DY others +jetstt

Diboson W+jets

QCD multijet Unc.

(13 TeV)-12016, 35.9 fb

CMSPreliminary

QCD ττ Control Regions

0jet vbfboosted

• ttbar control region produced from eμ channel• Affects rate in every

channel/category in fit• Pζ <-35

NOT MY

THESIS


Hadronic Tau Identification

75

• Working point efficiencies and misidentification probabilities


Status of 125 GeV Higgs

76

ATLAS and CMS continue Higgs measurements

in Run-II at LHC. ATLAS-CONF-2017-041CMS-PAS-HIG-16-043

• Does the Higgs have a Yukawa coupling? Further confirmation of the yukawa-coupling arrives.

• VH(bb) Run-II evidence at ATLAS with 3.5(3.0) σ observed (expected) with a signal strength µ =1.20 +0.24 -0.23 (stat) +0.34 -0.28 (syst) in 13 TeV. Combination with Run-1: σ observed (expected) 3.6(4.0) with μ = 0.90 ± 0.18(stat.) +0.21 −0.19(syst.)

• H(ττ) Run-II at 4.9(4.7) σ observed (expected) at CMS in 2016 data. (μ=1.06 +0.25 -0.24).

• How can we use it to find dark matter?


nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-13

013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

owU

nder

flow

80-9

090

-100

100-

110

110-

120

120-

1130

130-

140

140-

150

150-

160

Ove

rflow

Und

erflo

w80

-90

90-1

0010

0-11

011

0-12

012

0-11

3013

0-14

014

0-15

015

0-16

0O

verfl

ow

Exp.

(Obs

.-Exp

.)

1−0123

Even

ts/b

in

2−10

1−10

1

10

210

310

410 > 0 GeVττT

p > 100 GeVττT

p > 150 GeVττT

p > 200 GeVττT

p > 250 GeVττT

p > 300 GeVττT

p

Observed µτeτ→Z DY others +jetstt DibosonW+jets QCD multijet 125ττ→H Uncertainty

(13 TeV)-135.9 fbCMS Preliminary , Boostedµe

(GeV)ττmUnd

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Exp.

(Obs

.-Exp

.)

1−012

Even

ts/b

in2−10

1−10

1

10

210

310

410 > 0 GeVττT

p > 100 GeVττT

p > 170 GeVττT

p > 300 GeVττT

p

Observed hτhτ→Z DY others +jetstt DibosonW+jets QCD multijet 125ττ→H Uncertainty

(13 TeV)-135.9 fbCMS Preliminary , Boostedhτhτ

boosted postfit: eμ ττ

77

ττ boosted: mττ : Η pT

eμ boosted:mττ : Η pT

NOT MY

THESIS


nder

flow

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Und

erflo

w

95-1

15

115-

135

135-

155

Ove

rflow

Exp.

(Obs

.-Exp

.)

012

Even

ts/b

in

2−10

1−10

1

10

210



(13 TeV)-135.9 fbCMS Preliminary , VBFµe

(GeV)ττmUnd

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Und

erflo

w40

-60

60-7

070

-80

80-9

090

-100

100-

110

110-

120

120-

130

130-

150

150-

200

Ove

rflow

Exp.

(Obs

.-Exp

.)

1−0123

Even

ts/b

in2−10

1−10

1

10



(13 TeV)-135.9 fbCMS Preliminary , VBFhτhτ

vbf postfit: eμ ττ

78

ττ vbf: mττ : di-jet mass

eμ vbf:mττ : di-jet mass

NOT MY

THESIS

Ph.D. DefenseL. Dodd, U. Wisconsin Feb. 15 2018 (GeV)vismU

nder

flow

50-5

555

-60

60-6

565

-70

70-7

575

-80

80-8

585

-90

90-9

595

-100

Ove

rflow

Und

erflo

w50

-55

55-6

060

-65

65-7

070

-75

75-8

080

-85

85-9

090

-95

95-1

00O

verfl

owU

nder

flow

50-5

555

-60

60-6

565

-70

70-7

575

-80

80-8

585

-90

90-9

595

-100

Ove

rflow

Exp.

(Obs

.-Exp

.)

02

Even

ts/b

in

2−10

1−10

1

10

210

310

410) > 15 GeVµ(

Tp ) > 25 GeVµ(

Tp ) > 35 GeVµ(

Tp


(13 TeV)-135.9 fbCMS Preliminary , 0 jetµe

(GeV)ττm

0 100 200 300

Exp.

(Obs

.-Exp

.)

1−01

Even

ts/b

in

1

10

210

310

410


(13 TeV)-135.9 fbCMS Preliminary , 0 jethτhτ

0jet postfit: eμ ττ

79

ττ 0jet: 1-D mττ

eμ 0jet: mvis : μ pT

NOT MY

THESIS


Combined Impacts

80

302928272625242322212019181716151413121110987654321

θ∆)/0θ-θ(2− 1− 0 1 2

CMS_htt_tt_tt_vbf_13TeV_ZTT_bin_46

CMS_htt_mt_mt_vbf_13TeV_ZTT_bin_8

CMS_eFakeTau_1prong1pizero_13TeV

CMS_eFakeTau_1prong_13TeV

QCDScale_ggH

CMS_eff_trigger_tt_13TeV


CMS_htt_mt_mt_0jet_13TeV_W_bin_9

CMS_htt_mt_mt_0jet_13TeV_W_bin_33

CMS_htt_QCD_VBF_tt_13TeV

CMS_htt_et_et_vbf_13TeV_QCD_bin_19


CMS_htt_tt_tt_boosted_13TeV_ZTT_bin_32

CMS_scale_t_3prong_13TeV



CMS_scale_t_1prong_13TeV

CMS_eff_t_13TeV

CMS_htt_tt_tt_vbf_13TeV_QCD_bin_45

lumi_13TeV


CMS_htt_QCD_boosted_tt_13TeV

CMS_htt_zmm_norm_extrap_boosted_tt_13TeV


CMS_scale_t_1prong1pizero_13TeV

CMS_mFakeTau_1prong_13TeV

CMS_scale_met_clustered_13TeV

CMS_htt_mt_mt_vbf_13TeV_QCD_bin_18

CMS_scale_met_unclustered_13TeV

CMS_scale_gg_13TeV

CMS Internal

r∆0.05− 0 0.05

Pull Impactσ+1 Impactσ-1

• Highest impacts and respective pulls

• Combined all channels and all categories

• Theoretical uncertainty largest impact, unclustered energy scale second largest impact


Higgs Pt Uncertainty

81


pzeta definition

• Variable used to cut for top control region

• measure of MET collinearity with tau candidates

82


2D Boosted Category• Moved from svFit total pT to vectorial higgs pT

• Vectorial higgs pT distribution more smooth

• Small impact on expected limit: <2%.

83

p⌧⌧T = |~pLT + ~pL0

T + ~EmissT |

SVHiggspT

VectorialH

iggspT


et_vbf_RelativeStatHF0 2 4 6 8 10 12 14 16 18 20

1−10

1

10

210

310 (13 TeV)-135.9 fb

CMSUp 0.008, Down -0.022

W_CMS_scale_j_RelativeStatHF_13TeVUp

W

W_CMS_scale_j_RelativeStatHF_13TeVDown

Factorized JES Treatment • 27 sources of JES in all

categories and channels

• Poorly populated templates in sensitive regions caused impacts to behave poorly.

• JES shape treatment:

• Shapes either kept if variation is smooth, demoted to lnN if “spiky” template, or dropped from the channel and category if negligible.

84

W Template Up and Down “RelativeStatHF”

eτ vbf

example of poorly behaved template with

“spike”

example


spread the variationuse lnN

smooth shape“remove spike”

Use LnN

>=1 bin that changes by more than 10%, and total acceptance <2%

for up and down

yes

<10% of bins are different from

nominal template

smooth shapeUse LnN

yes

JES Treatment: Chart

85

up and down histograms are

exactly the same

uncertainty dropped

yes

no>=1 bin that

changes by more than 10%, and total acceptance >2% for up and down

>=1 bin where the relative change in acceptance is five

times the total relative change in

acceptance

no bin where the relative change in acceptance is five

times the total relative change in

acceptance

if if

keep shape

yes

the relative total change in

acceptance is less than 0.5% for the up and for the

down shapes

uncertainty dropped

yes

starthere!!

no no no


Standard Model (SM): Higgs I

86

scalar doublet

when μ2 < 0 and λ > 0

φ12+φ22=ν

MH2=2λv

v=246 GeV


MonoHbackup

87


Mono-Higgs Outlook

88

• Near future

• Produce more baryonic simulation samples

• Combine with other mono-Higgs decay channels

• Interpret results for DM-nucleon cross section exclusion

• Full Run-II

• Expand analysis to boosted regime where hadronic taus overlap with muon/electron to allow for overlapping tau topology

• Search for heavy scalar (ττ) + DM model

A thesis is never finished, only abandoned.—Marià Cepeda,

quoting Juan Alcaraz, quoting someone else


CMS Summary Invisible

89CMS-PAS-HIG-16-009 arXiv:1610.09218 [hep-ex]

https://arxiv.org/abs/1610.09218


Control Regions

90

(GeV)T,totM0-150 150-325 >325

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτμτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310

Observedττ→Z


bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτμτ


(GeV)T,totM0 200 400 600 800 1000 1200 1400 1600 1800 2000

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτμτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310



bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτμτ


(GeV)T,totM0-150 150-325 >325

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτhτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

Observedττ→Z

+jetsνν→ZW+jetsMultiboson/Single TopQCD/multijetSM 125 GeV Higgstt

bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτhτ


(GeV)T,totM0 200 400 600 800 1000 1200 1400 1600 1800 2000

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτeτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310

Observedττ→Z


bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτeτ


(GeV)T,totM0-150 150-325 >325

Obs

/Exp

0.5

1

1.5

(13 TeV)-1 2016, 35.9 fbhτeτ

L. DoddPh.D. Thesis

Even

ts

1

10

210

310Observed

ττ→Zll→Z

W+jetsMultiboson/Single TopQCD/multijetSM 125 GeV Higgstt

bkg. uncertainty

(13 TeV)-1 2016, 35.9 fbhτeτ


eτ same signμτ same sign

ττ same-sign

μτ W eτ W


Mono-Higgs Cutflow

91


Results II: Combination ττ+γγ

92

(GeV)Z`M10 210 310 410

) (fb

)χχ

H→ B

Z'

→(p

pσ

95%

C.L

.

2−10

1−10

1

10


Observed Limitττ

Observed Limitγγ

(GeV)Z`M310

) (fb

)γγχχ

→ A

H

→(p

pσ

95%

C.L

.

2−10

1−10

1

10


Observed Limitττ

Observed Limitγγ

Z’-2HDM Baryonic-Z’


Status of 125 GeV Higgs

93

Indirect limit on the branching fraction to Beyond the Standard Model (BSM)

decays, including dark matter (invisible decays)


• How does SM Higgs boson help with finding dark matter?

• With Run-I combination results, the indirect searches limit the beyond the standard model branching ratio of Higgs to 34%.

• Better H(ττ) and other measurements will help to constrain further.

• We can improve with direct searches.

https://arxiv.org/abs/1606.02266


Event Yield

94

background yields in sensitive regiondominated by statistical uncertainties


Experiment backup

95


Standard Model (SM): Higgs II

96

• Scalar boson discovered by CMS and ATLAS in 2012

• Significance of results driven by ZZ and γγ channels.

• Mass: 125 GeV measured spin-0 scalar

run-II goal: reduce uncertainties!



Proton Structure

97

• The LHC collides protons

• Partons (gluons and quarks) interact in collisions

• Gluons, valence quarks (u and d) and sea quarks (virtual q anti-q pairs)

• Parton distribution functions

• Deep inelastic scattering experiments (such as at HERA) provide the probability f(x) for a quark or gluon to have fraction of the protons momentum x

• Without accurate pdfs, no cross section normalization

gluons carry ~50% of proton momenta


Dynamic strip envelope

98


Solenoid

99

s

LR

track

3.8 T superconducting solenoid cooled to 4.7K

provides critical magnetic field to

measure momenta of particles


LHC Parameters

100


Radiation lengths

101


Interaction lengths

102


Dipole magnetic fields

103


Bethe-Bloch

104


(13 TeV)-1 2016, 36.8 fbhτµτ

CMSPreliminary

visible mass [GeV]0 20 40 60 80 100 120 140

Even

ts

0

10

20

30

40

50

60

70

310×Observed

±h

±

h±: hττ→Zs 0π ±: hττ→Z

±: hττ→Zµµ→Z

ElectroweakQCDtt

(13 TeV)-1 2016, 36.8 fbhτµτ

CMSPreliminary

(13 TeV)-1 2016, 36.8 fbhτµτ

CMSPreliminary

mass [GeV]hτ0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Even

ts

0

20

40

60

80

100

120

310×Observed

±h

±

h±: hττ→Zs 0π ±: hττ→Z

±: hττ→Zµµ→Z

ElectroweakQCDtt

(13 TeV)-1 2016, 36.8 fbhτµτ

CMSPreliminary

Hadronic Tau Reconstruction


SM Hττ 8TeV

106


SM Higgs TauTau 8 TeV Analysis

107

Straightforward mass-based analysis

Used reconstructed invariant di-Tau mass for signal extraction.

Multi-Variate Analysis(MVA) correction for MET based of the decay modes of each tau.

SVFit

Mass reconstructed using likelihood fit to determine most probable 4 vectors of taus, and met.

Assume all met in event is from taus and need to specify what decay mode of each tau for accurate neutrino estimates

tau decay modes τeτh , τμτh , τhτh , τeτμ , τeτe τμτμ

examined

WH/ZH additionSM H->ττ at 8TeV

CMS-HIG-13-004arXiv:1401.5041 [hep-ex]

http://cms-results.web.cern.ch/cms-results/public-results/publications/HIG-13-004/index.html


SM Higgs TauTau 8 TeV Analysis

108SM H->ττ at 8TeV

discriminatingvariables used

CMS-HIG-13-004arXiv:1401.5041 [hep-ex]

• 95% CL limits on SM H→ττ signal strength

• excess on mass spectrum from 115 GeV-135 Gev

and SVfit mass



Laura in HEP: summary

109

2012 2013 2014 2015 2016 2017

L1 Trigger MSSM Higgs->hh->ττ bb MSSM Higgs ->ττ SM h->ττ Mono-Higgs ττ

summer summerpreliminary exam

defenseand

find a job

2018

hardware supersymmetry: supersymmetry: thesis: thesis:


MSSM Hhh

110


SM Higgs as a tool for discovery

111 125 GeV H->ττ used as a tool for searches

• Di-Higgs 2 main flavors

• Measure: SM Trilinear Ηiggs coupling

• Search: New heavy resonance X

• Several di-Higgs bbττ analyses done at CMS

• Non-resonant

• Aims to measure trilinear Ηiggs coupling

• Resonant analysis

• Aims to reconstruct mass peak of “X”

• VH Analysis

• Search for resonant MSSM A->Zh(125)

Hhhττbb


bbττ

• Performed at 8 TeV

• 8 TeV Run1 in two different results

• MSSM-driven analysis and Model Independent Analysis

• Performed at 13TeV in 2015 and 2016

• 2016 Analysis has 12.9 /fb

• Resonant and non-resonant production modes

112

plot: CMS-PAS-HIG-16-011

BR h(bb) h(ττ)= 3.6%

125 GeV H->ττ used as a tool for searches

Hhhττbb

http://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/HIG-16-011/CMS-PAS-HIG-16-011_Figure_001-b.png


8 TeV bbττ

113

CMS-HIG-14-034CMS-PAS-HIG-15-013

HHKinFit

Maximum likelihood fit for mass Heavy Higgs, H, using the topology below.

Input: (1) SVfit rebuilds H->ττ first (2) 2 Reconstructed Jets

Di-jet invariant mass constrained to be 125 GeV

125 GeV H->ττ used as a tool for searches

• MSSM Analysis search for Heavy H decays to two 125 GeV h

• Probe low tanβ MSSM

• Model Independent analysis used same techniques and extended the mass range of the search

• Resonant and non-resonant

Hhhττbb


http://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/HIG-15-013/index.html


8 TeV MSSM H->hh->bbττ

114

CMS-HIG-14-034

• MSSM Analysis search for Heavy H decays to two 125 GeV h

• Kinematic fits assume no other MET in event

• Tau decay Modes: τeτh τμτh τhτh τeτμ

• τμτh most sensitive at low MH- similar to Legacy SM Higgs result.

• τhτh drives limit at high mass.

• Performed in tandem with A->Zh to probe low tanβ MSSM

Hhhττbb

ME: LTAU



8 TeV MSSM H->hh->bbττ

115 CMS-HIG-14-034

• A->Zh set stronger limits in MA-tanβ plane

Hhhττbb



8 TeV MSSM bbττ+llττ

116

+ =A->Zh H->hh COMB

CMS-HIG-14-034

Hhhττbb



MSSM Hττ

117


MSSM Higgs TauTau

118

MSSM Higgs Decays to ττ

CMS-PAS-HIG-16-006

• MSSM Ηiggs ττ search with 2.3 /fb in 2015 at 13 TeV

• Gluon-gluon fusion and associated-b production

• 2 categories: one b-tagged jet, no b-tagged jets using medium working point.

• Very similar analysis strategy and event selection to SM Higgs->ττ

• However, single lepton triggers used. No trigger requirement for the tau was applied.

• SVFit is used. The MT,tot variable rebuilt from svFit output performed better than Mττ variable



MSSM Higgs TauTau

119

CMS-PAS-HIG-16-006

• Event selection optimized for high mass

• MT,tot used to set limits

• Compared to 8 TeV better high mass limits, but not as sensitive to the low mass as 8 TeV, primarily due to better separation of ttbar backgrounds.

MSSM Higgs Decays to ττ


observation standard model higgs and the cms lhc€¦ · l. dodd, u. wisconsin feb. 15 2018 1 ph.d....

Documents