Inclusive Jet Production Cross Section using the KT Algorithm at CDF Run II PreBlessing Talk CDF Not

1
Inclusive Jet Production Cross Sectionusing the
KT Algorithm at CDF Run IIPre-Blessing Talk
CDF Note 7576

• R. Lefèvre, M. Martinez, O. Norniella
• QCD Meeting, April 8th 2005

2
Outlook

• Introduction
• Motivations
• Data samples / Event selection
• Trigger study
• MC simulation
• Data / MC comparison of raw variables
• Dijet Balance
• Bisector Method
• Jet PT Corrections
• Pile-Up Correction
• Absolute Correction
• Unfolding
• Systematic Uncertainties
• Underlying Event / Hadronization Correction
• NLO
• Results

3
Motivations

• Measure jet inclusive cross sectionfor central
  jets 0.1ltYlt0.7
• Stringent test of p-QCD
• Over more than 8 order of magnitudes
• Improve constraints on PDFs
• Tail sensitive to new physic
• Probing distances 10-19m
• Appreciate ?s enhancement for Run II
• KT preferred by theory
• Infrared and collinear safe to all orders in
  p-QCD
• No merging/splitting feature
• No RSEP issue comparing to p-QCD
• More sensitive to low PT contributions than cone
  based algorithms?
• Use 3 ? values for the D parameter
• D related to the size of the jets
• D 0.5, 0.7 and 1.0

4
Data samples / Event selection

• Framework
• Version 5.3.3nt of the code latest primary
  vertex finder algorithm
• Jet datasets xxxx0d latest calorimeter
  calibration (13A)
• Version 5.3.3 of the Monte-Carlo
• Pythia Tune A taken as nominal MC, Herwig only
  considered for systematics
• Run selection
• Version 7 of QCD good run list (no Silicon
  requirement)
• Run 155368, 155742 excluded discard 3.8 pb-1
• Cross section dropped of about 40
• Integrated luminosity 385 pb-1 (1.019 corrective
  factor take into account)
• Event selection
• Jets defined with the KT algorithm D 0.5, 0.7
  or 1.0
• At least one jet with 0.1 lt YJETlt0.7
• At least one primary vertex of Quality ?12, best
  primary vertex Vz lt 60 cm

5
Cut study
6
Sanity checks
7
Trigger study method

• Trigger Structure
• L1, L2 and L3 trigger efficiencies extracted from
  data
• Minbias events efficiency of Stw5 (L1)
• STW5 events efficiencies of J15 (L2) and J20
  (L3)
• JET20 events efficiencies of Stw10 (L1), J40
  (L2) and J50 (L3)
• JET50 events efficiencies of J60 (L2) and J70
  (L3)
• JET70 events efficiencies of J90 (L2) and J100
  (L3)
• Trigger thresholds efficiency (L1?L2?L3) ? 99
• To avoid systematics due to energy scale
  uncertainties, the obtained thresholds are
  increased by 5 for final results

8
Trigger study results
Trigger efficiency gt 99 raw PT in GeV/c
9
Raw cross section
Check consistency at overlap regions
Each event weighted by its own prescale
10
MC simulation

• CDF simulation
• Tracks E / p reasonably well reproducedfor
  central calorimeters
• Pythia Tune A used as nominal MC
• Includes tuned parametersfor the Underlying
  Event
• Reproduces the Jet Shapes
• Outlook
• Data / MC comparison of raw variables
• Dijet Balance
• Bisector Method
• Additional requirement
• 1 and only 1 primary vertex of Quality ?12

11
Comparison of raw variables
MC black
Data blue
12
Dijet Balance method

• Event selection
• 2 and only 2 jets with PTRAW ? 10 GeV/v
• One jet (trigger jet) with 0.2 lt ?DTRIG lt 0.6
• The other jet (probe jet) with 0.1 lt YJETlt0.7
• 1 and only 1 primary vertex of Quality ?12 ,
  Vz lt 60 cm
• Missing ET significance criterion apply using
  lowest PTRAW jetto set the effective cut
• Definitions
• PTMEAN (PTPROB PTTRIG) / 2
• ?PTF (PTPROB - PTTRIG) / PTMEAN
• In bin of PTMEAN
• ? (2 lt?PTFgt) / (2 - lt?PTFgt)
• Event by event ? PTPROB / PTTRIG

13
Dijet Balance results
14
Bisector Method

• Event selection
• 2 and only 2 jets with PTRAW ? 10 GeV/v
• The 2 jets with 0.1 lt YJETlt0.7
• 1 and only 1 primary vertex of Quality ?12 ,
  Vz lt 60 cm
• Missing ET significance criterion apply using
  lowest PTRAW jetto set the effective cut
• Definitions
• PTMEAN (PTRAW1 PTRAW2) / 2
• ? (?JET1 - ?JET2) / 2
• ?PT// (PTRAW1 PTRAW2) cos(?)
• ?PTPERP (PTRAW1 - PTRAW2) sin(?)
• In bin of PTMEAN
• ?// rms of ?PT// distribution
• ?PERP rms of ?PTPERP distribution
• ?D ? (?2PERP - ?2//) / ? 2

15
Bisector Method results
16
Pile-Up Correction method

• Correction
• PTRAW (Pile-Up Corrected) PTRAW ?D ? (NVQ12
  1)
• ?D extracted from the data
• Shapes of normalized cross sections vs PTRAW
  (Pile-Up Corrected) dividing the data in 2
  sub-samples of instantaneous luminosity
• High Luminosity / Low Luminosity
• Low Luminosity 5 to 15 ? 1030 cm-2s-1
• High Luminosity gt 35 ? 1030 cm-2s-1

17
Pile-Up Correction results
Compatible results obtained using other high
luminosity sub-samples
18
Pile-Up Correction MC check method

• Ratio of cross sections vs PTRAW (Pile-Up
  Corrected) as obtained with 1 pile-up MC and no
  pile-up MC
• 1 Pile-Up MC / No Pile-Up MC

19
Pile-Up Correction MC check results
Results compatible with the one obtained from the
data
20
Absolute correction method

• Method
• KT algorithm run at calorimeter and hadron level
• Pair of calorimeter-hadron jets matched in the
  Y- ? space
• ?R ? (?Y2 ??2) lt D
• Closest hadron jet if more than one within ?R
  requirement
• ltPTHAD PTRAWgt vs ltPTRAWgt
• Fit by a 4th order polynomial
• 2 ways
• In PTRAW bins
• In (PTHAD PTRAW) / 2 bins

21
Absolute correction results
PTRAW bins
(PTHAD PTRAW) / 2 bins
22
Unfolding method

• MC
• Jets at hadron level
• No cut but YJET applied on hadron level jets
• NiHAD
• Jets at calorimeter level
• All cuts applied, use PTCOR
• NiCAL
• Bin-by-bin unfolding factor Ci NiHAD / NiCAL
• Data using PTCOR
• NiDATA UNFOLDED Ci ? NiDATA NOT UNFOLDED

23
Unfolding choice of absolute correction
PTRAW bins
(PTHAD PTRAW) / 2 bins
24
Re-weighting Pythia method

• Ratio Data / Pythia
• Fit by a 3rd order polynomial
• Pythia re-weight applying this 3rd order
  polynomial to pT hat

25
Re-weighting Pythia results
26
Unfolding using Herwig
27
Systematic uncertainties (1/2)
28
Systematic uncertainties (2/2)
29
UE / Hadronization correction
?

?
30
CHAD vs D

• Correction limited to PTJET gt 54 GeV/c

31
NLO

• JETRAD CTEQ61 package
• ?R ?F Maximum Jet PT / 2
• K-factor (NLO / LO) 1 for 70GeV/c
  hep-ph/0303013
• NLO uncertainties
• Scale ?R ?F Maximum Jet PT
• Symmetric uncertainties
• PDF
• Asymmetric uncertainties
• Dominates by gluon at high-x contribution
• PDF uncertainties dominate
• NLO corrected to hadron level multiplying the
  prediction by CHAD

32
Results D0.5
33
Results D0.7
34
Results D1.0
35
Conclusion

• PTJET from 54 to 800 GeV/c
• Measurements extend over 8 orders of magnitude
• Very good agreement with NLOover all the PTJET
  range
• Excess at very high PTJET not statistically
  significant

36
Event Display 2nd highest PTJET event
37
Event Display highest PTJET event