Tau Energy Scale in Tau Energy Scale in 2016 dat 2016 dataset aset Huajie Cheng 2020.11.13
Introduction ◆ TES measurement using TauFW ◼ Preliminary results using 2016 dataset, in mu-tau events ◼ Coarse ES variations applied to real tau • -3% → 3% in steps of 1% (0.2% for final measurement) ◼ Compare the impacts in DY and TT processes ◼ Fit independently per DM • Only in DY process • Use m vis and m τ as observables • Fix “r” parameter and scan likelihood profile in TES • Fit the NLL profile by asymmetric parabola function 2
Mu-tau selections ◆ Baseline ➢ Single muon trigger & SF applied ➢ Muon • pT > 23 GeV, |eta| < 2.1 or 2.4 • |dxy| < 0.045, |dz| < 0.2 • medium ID, rel. iso < 0.15 ➢ Tau • pT > 20 GeV, |eta| < 2.3, |dz|<0.2, decay modes 0, 1, 10, 11 • DeepTau2017v2p1VSjet Tight WP • DeepTau2017v2p1VSmu Tight WP • DeepTau2017v2p1VSe VVLoose WP ➢ Opposite sign mu- tau pair with ΔR > 0.5 and highest mu and tau pT ➢ Extra-lepton veto ◆ m T < 50 GeV 3
TES systematic uncertainties ◆ All the shape uncertainties are not yet in place Nuisance Parameter Distribution Uncertainty Applied to ± 2.5% lumi lnN all, except QCD ± 2% mu Eff lnN all, except QCD tau ID shape from recommendation ZTT, TTT ± 2% DY cross section lnN DY ± 6% ttbar cross section lnN ttbar ± 5% signle top cross section lnN single top ± 5% VV cross section lnN diboson ± 8% W+jets normalization lnN W ± 10% QCD normalization lnN QCD ± 15% jet -> 𝜐 fake rate lnN ZJ, W, QCD, TTJ, ST ± 5% on jet → 𝜐 energy jet -> 𝜐 fake ES shape ZJ, W, TTJ ℓ -> 𝜐 fake rate shape from recommendation ZL, TTL ± 2% on ℓ → 𝜐 energy ℓ -> 𝜐 fake ES shape ZL, TTL apply weight ± 10% Z pt reweight shape DY 4
Datacard for DM0 ◆ Not all syst included, e.g. tid 5
Likelihood profile scan ◆ Profile the NLL as a function of tes ◆ And the fit with asym parabola function DM1 DM10 DM11 DM0 DM1 DM10 DM11 6
Fit result ◆ DM1 for m_tau DM1 7
Fit result ◆ DM10 for m_tau DM10 8
Fit result ◆ DM11 for m_tau DM11 9
Fit result ◆ DM0 for m_vis DM0 10
Fit result ◆ DM1 for m_vis DM1 11
Fit result ◆ DM10 for m_vis DM10 12
Fit result ◆ DM11 for m_vis DM11 13
Results ◆ Finally got the TES values w.r.t different DMs Before fit After fit Ref. ◆ Quite ugly due to very coarse variation step and missing uncertainties ◆ Jobs with step=0.2% almost done! ◆ Asked to give a status report next Monday 14
Some distributions -3% TES +3% TES ◆ Here are the plots with all DMs combined, the fit will be performed with each DM separately 16
Some distributions Negligible impact from ttbar ◆ Will only apply the TES in ZTT for the measurement 17
Framework: from nanoAOD to final plots 18
Sample list 19
Air condition monitoring in the clean room ◆ Took 4 samples in each rooms (2 for room 110) 10000 0.3um 0.5um 1.0um 2.0um 5.0um 10.0um Counts in 5 mins (14.16L) 1000 100 10 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Cases
IV-curve ◆ Finally managed to perform IV measurement remotely by LabView ➢ Change from GPIB to RS232-to-USB cable
Recommend
More recommend
