Top Mass Measurements at the CDF Experiment

1
Top Mass Measurements at the CDF Experiment

• Luca Scodellaro
• Instituto de Fisica de Cantabria
• for the CDF Collaboration

QCD 05 12th International QCD Conference 4-9th
July, Montpellier (France)
2
Outline

• Motivations for measuring the top mass
• Top production at the Tevatron
• Top mass measurements at CDF
• Summary and Perspectives

Measurement Challenges b quark tagging Jet energy
corrections
Template Method Matrix Element Method
3
Motivations for Measuring Top Mass

• Special Role in SM due to the high mass
• Precise top mass measurement provides

Dominant parameter in radiative corrections
for Electroweak predictions
Consistency tests for the Standard
Model Constraint on unknown parameter (Higgs
boson mass) Sensitivity to physics beyond the
Standard Model
4
Top Production at the Tevatron
_

• Tevatron pp collisions at vs1.96 TeV
• Pair production is the
• dominant mechanism
• BR(t-gtWb) 100
• Observed final states given by W boson decays

Dilepton tt -gt l?l?bb (5) Leptonjets
tt -gt lvqqbb (30) Hadronic tt -gt qqqqbb
(45)
lelectron or muon
5
Measurement Challenges

• LeptonJets channel
• Neutrino not detected
• Px and Py from transverse energy
  conservation
• Pz from MwMlv constrain (two solutions)
• Four expected jets in the final states
• Ten possible choices for parton-jet
  assignments
• Dilepton channel
• Low statistics
• Two neutrinos in final states
  under-constrained kinematic system
• Jet Energy Measurement
• Largest contribution to top mass uncertainty

6
b Quark Tagging

• Tagging the b quark improves parton-jet
  assignments
• leptonjets 1 tag-gt 6 combinations, 2 tags-gt2
  combinations
• Requiring a b tagged jet also reduces background
  contamination
• B hadrons are long lived
• Jet tagging
• efficiency -gt 42
• False tag rate -gt 0.5

7
Jet Energy Corrections

• Big effort in understanding jet energy
  measurement
• 3 jet pT uncertainty
• in top events
• Big improvements
• for plug jets (?gt1)
• with respect to RunI
• due to new detector
• Significant improvement with respect 2004
  analyses

Central Jets (?lt1)
-- RunII -- RunI -- Winter 2004
8
The Template Method

• Reconstruct Mtop per event
• kinematic fit to determine the most probable
  jet-parton assignment
• Top mass measurement
• best likelihood fit of mass distribution in
  data to background and signal templates for
  different top mass values

9
LeptonJets Jet Energy Scale Fit

• W mass reconstruction used to fit the JES
• Mass templates for top and W as a function of top
  mass and JES
• Data divided in 0-tag, 1-tag and 2-tags sub
  samples
• Fit data simultaneously for Mtop and JES
• 20 improvement in uncertainty due to JES
  

Mtop 173.5 2.7 -2.6 (stat.)2.5(JES)1.7(syst.
) GeV/c2
Most precise single top mass measurement in the
world
10
LeptonJets Additional b Tagging

• Increasing S/B ratio and parton-jet assignment
• by using further b tagging techniques

The Jet Probability algorithm computes the
probability for a set of tracks inside a jet to
come from the primary interaction of the event
Mtop 173.0 2.9 -2.8 (stat.)3.3(syst.) GeV/c2
11
Template Method in Dilepton Events

• Two neutrinos in the final states
• additional assumptions to constrain the event
  kinematic
• Neutrino weighting approach
• Mtop maximizing probability W used to build
  templates

Probability for assumed neutrinos ?
Assumed theoretical value of MW
W(Mtop,Mw)
Missing ET constraints
Integration over parton assignments and
neutrinos ?
Mtop 170.6 7.1 -6.6 (stat.) 4.4 (syst.)
GeV/c2
12
Matrix Element Method

• Not to look only to a particular parton-jet
  assignment
• Compute a probability as a function of Mtop
• Sum over all the unknown in the event
• parton-jet assignments and neutrino solutions
• Top mass from maximum likelihood fit

Transfer functions connect parton to jet variables
Matrix element
Parton distribution function
L ?iP(xiMtop)
13
Matrix Elements in LeptonJets Events

• Only signal probability considered
• Background effect described
• by a mapping function
• obtained by pseudo-experiments
• for different background fractions

• One inclusive b tag required
• Mapping function at 21 applied

Mtop 173.8 2.7 -2.5 (stat.)3.3(syst.) GeV/c2
14
Matrix Elements in Dilepton Events

• Background probabilities included in the
  likelihood
• P(xi Mtop) psPs(xi Mtop)
  SjpbgjPbgj(xi)

• 33 candidate events
• observed in 340 pb-1
• of data
• Top Mass measured
• from the expectation
• value of the posterior
• probability density

Mtop 165.3 6.3 (stat.) 3.6 (syst.) GeV/c2
15
Summary and Perspectives

• TevEWWG has produced a new average
• Mtop 174.33.4 GeV/c2
• Currently using most precise single measurement
  per decay channel from each experiment
• Working ongoing to combine results within each
  channel (including CDF results from RunI)

16
Summary and Perspectives

• Electroweak fit to Higgs boson mass using latest
  Tevatron combination
• New central value
• MH9852-36GeV/c2
• New upper limit
• MHlt208GeV/c2 _at_95CL

17
Summary and Perspectives

• Fitting the jet energy scale from W decays will
  reduce systematic uncertainty with luminosity
• Uncertainty on Jet Energy Scale less than 1
  GeV achievable
• Can reach our RunII goal ?Mtop 2-3GeV/c2

18
Backup Slides
19
The Tevatron Collider

• Tevatron RunII
• pp at vs1.96 TeV
• Peak luminosity
• L 1.21032

_
20
The CDF Detector
21
Electroweak fits
22
Comparison with EW Fits