Top quak mass measurement using m T2 at CDF (dilepton channel) - - PowerPoint PPT Presentation

▶
top quak mass measurement using m t2 at cdf
SMART_READER_LITE
LIVE PREVIEW

Top quak mass measurement using m T2 at CDF (dilepton channel) - - PowerPoint PPT Presentation

Feb 02, 2023 402 likes •582 views

Top quak mass measurement using m T2 at CDF (dilepton channel) Hyunsu Lee The University of Chicago On behalf of the CDF collaboration PHENO 2009 Symposium Hyunsu Lee, The University of Chicago Why we measure the top mass in the dilepton

slide-1
SLIDE 1

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Top quak mass measurement using mT2 at CDF

(dilepton channel) Hyunsu Lee The University of Chicago On behalf of the CDF collaboration

slide-2
SLIDE 2

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Why we measure the top mass in the dilepton channel

  • It is important to check the mass

crossing the channels

Is it SM top? Significant difference indicate the new physics

  • This channel can be a standard

candle for new physics search

Well known SM process Signal and background is under control Similar topology with new physics Pair produced new particle can have two missing particle final state

2

slide-3
SLIDE 3

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Signature of new physics particle and mT2

  • New physics predict the candidate
  • f dark matter (WIMP)

Ex) Neutralino in the SUSY

  • If we consider pair production of

new physics particle

Two missing particle Visible particle (quark and leptons) Ex) two gluino pair production

  • We are interesting to determine the

mass of new particle

mT2 was introduced for two missing particle

m T2 = m in[m ax(m T(1),m T(2))] qT+p T=m issing pT

slide-4
SLIDE 4

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

mT2

  • Transverse mass of two missing

particle system

Similar with mT for W mass

  • Can be useful to determine the

mass of new physics particle

One of the most stringent variable

  • Top dilepton channel is good

example of mT2 variable (standard candle)

  • We can use real data

First application in the real data

)

2

mT2 (GeV/c 50 100 150 200 250 300 Arb units 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

Tagged DIL

)

2

= 160 (GeV/c

top

M )

2

= 170 (GeV/c

top

M )

2

= 180 (GeV/c

top

M

Tagged DIL

4

slide-5
SLIDE 5

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

ttbar dilepton channel

  • Dilepton (5% branching

ratio, small background) 2 high-PT leptons(e/m), 2 b-jets, large missing ET

We separate subsample with b tagging

slide-6
SLIDE 6

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

6

Template method

MC

tt backgrounds

Event reconstruction 2d templates mt

reco+ mT2

mt

reco+HT

1d templates mT2 mt

reco

HT DATA Event reconstruction Likelihood Fit

slide-7
SLIDE 7

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Sanity check

)

2

(GeV/c

top

M 160 165 170 175 180 )

2

residual (GeV/c

  • 2
  • 1.5
  • 1
  • 0.5

0.5 1 1.5 2

T2

Residuals: Dilepton mt+m

Constant 0.08797 ± 0.08424 Constant 0.08797 ± 0.08424

  • 1

CDF II preliminary 3.4 fb

T2

Residuals: Dilepton mt+m

)

2

(GeV/c

top

M 160 165 170 175 180 pull width 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2

T2

Pull widths: Dilepton mt+m

Constant 0.005208 ± 1.006 Constant 0.005208 ± 1.006

  • 1

CDF II preliminary 3.4 fb

T2

Pull widths: Dilepton mt+m

)

2

(GeV/c

top

M 160 165 170 175 180 )

2

residual (GeV/c

  • 2
  • 1.5
  • 1
  • 0.5

0.5 1 1.5 2Residuals: Dilepton mT2 only

Constant 0.1047 ±

  • 0.2626

Constant 0.1047 ±

  • 0.2626
  • 1

CDF II preliminary 3.4 fb

Residuals: Dilepton mT2 only

)

2

(GeV/c

top

M 160 165 170 175 180 pull width 0.85 0.9 0.95 1 1.05 1.1 1.15 1.2 Pull widths: Dilepton mT2 only Constant 0.005466 ± 0.9969 Constant 0.005466 ± 0.9969

  • 1

CDF II preliminary 3.4 fb

Pull widths: Dilepton mT2 only

slide-8
SLIDE 8

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Expected statistical uncertainties

8

slide-9
SLIDE 9

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Systematics and total estimated uncertainty

175 GeV/c2 top mass assumed

9

mT2 give the best performance between single observables

Unit (GeV/ c2)

slide-10
SLIDE 10

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Data fit result

169.3 ±2.7 (stat.) ± 3.2 GeV/c2(syst.) 168.0 (stat.) ± 2.9 GeV/c2(syst)

  • 4.0

+4.8

m T2 and m t

NWA

m T2 alone

10

slide-11
SLIDE 11

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Data distribution

)

2

(GeV/c

NWA t

m 100 150 200 250 300 350 )

2

Events/(10 GeV/c 2 4 6 8 10 12 14 16 18 20 22 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

)

2

(GeV/c

NWA t

m 100 150 200 250 300 350 )

2

Events/(10 GeV/c 5 10 15 20 25 30 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

)

2

(GeV/c

T2

m 50 100 150 200 250 300 )

2

Events/(10 GeV/c 5 10 15 20 25 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

)

2

(GeV/c

T2

m 50 100 150 200 250 300 )

2

Events/(10 GeV/c 2 4 6 8 10 12 14 16 18 20 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

(GeV/c)

t

H 200 300 400 500 600 700 800 Events/(30 GeV/c) 2 4 6 8 10 12 14 16 18 20 22 24 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

)

2

(GeV/c

t

H 200 300 400 500 600 700 800 Events/(30 GeV/c) 5 10 15 20 25 30 Data Signal+Bkgd Bkgd only

)

  • 1

CDF II Preliminary (3.4 fb

Non-tagged m t

NWA

Non-tagged m T2 Non-tagged H T tagged m t

NWA

tagged m T2 tagged H T

11

slide-12
SLIDE 12

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Conclusion

  • We measure the top quark mass in the dilepton channel

using mT2 variable

169.3 ±2.7 (stat.) ± 3.2 GeV/c2(syst.) with mt

NWA and mT2

168.0 (stat.) ± 2.9 GeV/c2(syst) with mT2 alone First application in the real data

  • We prove the performance of mT2

Best single observable including systematic

  • This method can be useful to mass determination of new

physics particle

12

  • 4.0

+4.8

slide-13
SLIDE 13

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

mtNWA

  • Leptonical decay of top

t->blv We measure b and lepton but don’t know neutrino 4 unknown Known parameter W mass neutrino mass (2 unknown) If we assume the top quark mass and neutrino eta direction, we can measure neutrino x,y momentum Same thing happen for the

  • ther leg
  • Getting weight using

measured missing transverse energy

t W b l ν

1 2

( , , )

i i top

w w m

ν ν

η η =

13

slide-14
SLIDE 14

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

mtNWA

  • Some over neutrino rapidities
  • We have maximum weight mt

as reconstructed mass (mt

NWA)

  • We scan mt with 3GeV size

and then decrease the step size upto 0.15GeV near the peak

  • We have gaussian fit in the

near of peak to get mt continuously

j i j i t t

m w η P η P dη dη = m W

,

) ( ) ( ) ( ) (

∑∑ ∫ ∫

2 1 2 1

14

slide-15
SLIDE 15

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Ht

  • Linear sum of jets, leptons, and missing Et
  • Strong JES correlation and also strong Top Mass

correlation (strong correlation with mt

NWA)

15

slide-16
SLIDE 16

PHENO 2009 Symposium Hyunsu Lee, The University of Chicago

Why we measure top quark mass

  • SM Higgs Mass was

constrained by Mtop and MW through loop correction of W mass

  • Precision top quark mass

measurement

Predict SM Higgs mass Constraints for physics beyond standard model

X ??

16