0 tt pp x
play

0 tt pp X Yuri Oksuzian University of Florida PHENO 2010 1 Why - PowerPoint PPT Presentation

Oct 25, 2023 •353 likes •682 views

Resonance search in 0 tt pp X Yuri Oksuzian University of Florida PHENO 2010 1 Why and How? t Goal is to test production for possible new sources t such as a narrow resonance Top is very heavy, maybe indication of

  1. Resonance search in 0 → tt pp → X Yuri Oksuzian University of Florida PHENO 2010 1

  2. Why and How? t ¯ • Goal is to test production for possible new sources t such as a narrow resonance  Top is very heavy, maybe indication of coupling to new physics  Top is a young particle  Various theoretical models predict it: technicolor, KK gluons • Search technique:  M tt spectrum is reconstructed, using FlaME  Search for a peak in M tt spectrum – Understand SM fluctuation probabilities – Calculate UL(Upper Limits) – Compare data with our expectations(SM or with new physics) PHENO 2010 2

  3. Where? This is the first M tt analyses in All Hadronic channel •  Disadvantages – Large QCD background » Controlled with good event selection – More combinations  Advantages – Highest branching ratio t ¯ » Most events are here t – No missing information like neutrino » Better signal templates  Future – Combined result with lepton+jets channel » Higher sensitivity – Cross-check for a possible discovery PHENO 2010 3

  4. Motivation - previous results Better agreement Excess ~500Gev with SM :( PHENO 2010 4

  5. FlaME (Florida Matrix Element) We calculate the a priori probability density for an event to be the result of Standard Model production and decay t ¯ t 1 σ ( M top ) ε ( M top ) N combi Σ P ( j | M top ) = dz b dz b f ( z a ) f ( z b ) d σ ( M top , p ) TF ( j | p ) P T ( p ) ∫ combi To calculate the M tt probability density, we modify the integral above: 1 σ ( M top ) ε ( M top ) N combi Σ ρ ( x | j ) = dz b dz b f ( z a ) f ( z b ) d σ ( M top , p ) TF ( j | p ) P T ( p ) δ ( x − M tt ( p )) ∫ combi As M tt estimator we use average of this distribution: M tt =< ρ ( x | j ) > PHENO 2010 5

  6. MC/Data Samples • Signal samples: t ¯  Pythia generated narrow resonant samples t with masses 450, 500 ... 900 GeV • Background Samples: Signal Templates CDF Run II MC preliminary 0.14 All Hadronic t ¯  SM MC sample Lepton + jets t 0.12  QCD 500 0.1 – Data driven 700 0.08 0.06 900 0.04 0.02 0 300 400 500 600 700 800 900 1000 1100 1200 2 M [GeV/c ] tt PHENO 2010 6

  7. Trigger & Prerequisites • Multijet Trigger  L1: ≥ 1 tower with E T ≥ 10 GeV  L2: ≥ 4 clusters with E T cl ≥ 15 GeV, Σ E T ≥ 125 GeV  L3: N jet ≥ 4, with E T jet ≥ 10 GeV – σ ≈ 14 nb, ~85% all hadronic efficiency • Prerequisites  Good run list  Vertex: |z|<60cm & |z-z p |<5cm  Missing Et Significance: < 3 (GeV) 1/2  Tight lepton veto  6,7 tight jets - E T jet ≥ 15GeV, | η |<2.0 t ¯ • After prerequisites we have /QCD~ 1/1000! t PHENO 2010 7

  8. Neural Net Idea Neural net event selection: •  Uses a Root class TMultiLayerPerceptron  11 inputs, 2 hidden layers with 20/10 nodes and 1 output SumEt - total transverse energy • ∑ E T − E T 1 − E T 2 SumEt3 - sub-leading transverse energy • ∑ ˆ C - centrality: E T / s • r j P j P j ∑ ∑ M ab = P / A - aplanarity: 3/2*(smallest eigenvalue) of • a b j j E* N - geom average of transverse energy of the N-(2 leading jets) • E* T1 - transverse energy of the leading jet • CDF Run II MC preliminary M 2j min - the minimum dijet mass • M 2j max - the maximum dijet mass • M 3j min - the minimum trijet mass • M 3j max - the maximum trijet mass • FlaME variable, ∑ -Log(P( M top =155,160…195GeV)) • PHENO 2010 8

  9. QCD background We build tag matrix from events from 4,5 jet events. • Each element in the matrix defined as: • The probability to single/double tag an event: • We weight each event in pre-tagged data sample to get the prediction • for 1, 2 tagged events Finally, we define several control region and test our modeling with • observation For all control regions we get a very good agreement • Biggest impact on final result comes from possible signal contamination, • using this procedure PHENO 2010 9

  10. Crosscheck with data 0.75<NNetOut<0.93. Mtt Mtt Mtt CDF Run 2 preliminary CDF Run 2 preliminary 3 QCD (Data-Model)/Model QCD + SMtt 1600 1600 Chi2/NDF=25.7/39 1,2 tag data 2 prob=0.846 SMtt, Norm to Data 2 2 1400 1400 ! ! / ndf / ndf 22.18 / 42 22.18 / 42 Z’(700GeV), Norm Prob Prob 0.9949 0.9949 Chi2/NDF=25.4/39 p0 p0 0.05098 0.05098 ± ± 0.09418 0.09418 1200 1200 1 prob=0.855 p1 p1 -7.727e-05 -7.727e-05 ± ± 2.008e-04 2.008e-04 1000 1000 0 800 800 600 600 -1 400 400 -2 200 200 -3 0 0 200 200 400 400 600 600 800 800 1000 1000 1200 1200 1400 1400 200 400 600 800 1000 1200 1400 PHENO 2010 10

  11. Limit Setting Methodology Template event weighting •  N X0 : based on assumed cross-section and acceptance  N tt : based on theoretical cross-section and acceptance  N QCD : Balance from data tot = N cdf Ldt ⋅ ( σ X 0 A X 0 + σ tt A tt ) + N QCD ∫ Likelihood •  N X0 , N tt , N QCD are used to compute the expected number of events in mass bin “i”: µ ( i ) = N X 0 T X 0 ( i ) + N tt T tt ( i ) + N QCD T QCD ( i )  Given the observed number of events n(i) and expected µ (i) in bin “i”, the likelihood is equal to: n i L ( σ X 0 , r ν | r µ i e − µ i ∏ n ) = n i ! PHENO 2010 11

  12. Posterior density function Acceptance uncertainties accounting • d r ν ⋅ L ( σ X 0 , r n ) ⋅ π ( σ , r p ( σ X 0 , r ν | r n ) = ∫ ν ) We integrate over the nuisance parameters, uncertainties for: •  Signal acceptance  Background acceptance  Background cross-section Given p( σ |n) we define: •  σ X0 - max of PDF  95% confidence level upper UL 1 p ( σ | r limit(UL) n ) d σ = 0.95 ∫ Area 0  Values are calculated as median after 1000 PE’s PHENO 2010 12

  13. Systematics -1 CDF Run 2 preliminary, L=2.8fb [pb] To consider systematics, Xo Mass 450 • Xo Mass 500 1 which both affect shape Xo Mass 550 Xo Xo Mass 600 ! and acceptances, we: " Xo Mass 650 Xo Mass 700 Consider the shift on 0.8 • Xo Mass 750 Xo Mass 800 cross-section by: Xo Mass 850 Xo Mass 900 Running PE from shifted • 0.6 templates and fit them with nominal ones 0.4 We considered • systematics due to JES, 0.2 ISR/FSR. PDF found to be negligible 0 0 0 0.5 0.5 1 1 1.5 1.5 2 2 2.5 2.5 3 3 3.5 3.5 4 4 [pb] [pb] ! ! Xo Xo PHENO 2010 13

  14. Applying systematics � ∞ � � 2 � 1 − 1 � σ X 0 − σ ′ PDF ( σ ′ ) · d σ ′ PDF SY S ( σ X 0 ) = 2 π exp √ 2 δσ X 0 δσ X 0 0 Cross-section posterior p.d.f. Cross-section posterior p.d.f. CDF preliminary -28 x10 likelihood 0.25 0.2 < 4.220 at 95% CL ! 0.15 < 4.420 at 95% CL ! 0.1 0.05 0 0 1 2 3 4 5 6 7 8 , pb ! Xo APS April Meeting 2009 14

  15. Data/BG prediction -1 -1 CDF Run II preliminary, L=2.8fb CDF Run II preliminary, L=2.8fb 2 2 500 500 events/20GeV/c events/20GeV/c QCD QCD SM t t SM t t CDF data, Nev=2086 CDF data, Nev=2086 2 2 400 400 10 10 300 300 10 10 200 200 100 100 1 1 0 0 300 300 300 400 400 400 500 500 500 600 600 600 700 700 700 800 800 800 900 900 900 1000 1000 1000 300 300 300 400 400 400 500 500 500 600 600 600 700 700 700 800 800 800 900 900 900 1000 1000 1000 2 2 2 2 M M [GeV/c [GeV/c ] ] M M [GeV/c [GeV/c ] ] tt tt tt tt PHENO 2010 15

  16. Upper Limits 4 ) [pb] Expected limit at 95% C.L. t 3.5 Expected limit at 95% C.L. ± 1 # t ! 0 Expected limit at 95% C.L. 2 ± # BR(X 3 Observed limit at 95% C.L. " Xo 2.5 Leptophobic Z’, $ =1.2% M Z’ Z’ # 2 1.5 1 0.5 450 500 550 600 650 700 750 800 850 900 2 M [GeV/c ] Xo PHENO 2010 16

  17. Conclusions&Plans First search for narrow ttbar resonance in all jets final state •  No excess found in 2.8/fb of CDF data  We set observed upper limit on leptophobic Z’ mass up to 805 GeV  Used various tools to test SM for very small % "contaminations of new physics" Analysis has been reviewed and no unresolved issues • were found Plan •  PRD publication  Updated result with improvements in lepton plus jets PHENO 2010 17

  18. Backup slides PHENO 2010 18

  19. Signal contamination • Signal contribution to QCD shape will be treated as following: From Equtaion 4 we have, number of events in bin “i”: µ = σ s A s T s + σ tt A tt T tt + N pure QCD T pure QCD N pure QCD T pure QCD = N cont QCD T cont QCD − σ s A cont T cont − σ tt A cont T cont s s tt tt Comparing signal templates of predicted and observed values we can assume: T s = T cont s So, finally we get: µ = σ s ( A s − A cont ) T s + σ tt A tt T tt + N cont QCD T cont QCD − σ tt A cont T cont s tt tt • In the end, it decreases signal acceptances by the values we get from TRM, which is about 1-1.5% • It will obviously result in the worse sensitivity. PHENO 2010 19

  20. Simplifications • To calculate that probability we need to compute 28 integrals:  P t and P z of incoming partons  4-momenta of 6 final partons • To reduce CPU time, we made some assumptions:  Pt of the incoming partons is 0. -2 integrals  All quarks except top are massless. -8 integrals  Partons and jets have the same direction. -12 integrals  W’s and top’s are on shell. -4 integrals • Only 2 integrals in total. We’ll do more in the future. PHENO 2010 20

Recommend

More recommend