Home

#Welcome to the ZInvisible wiki!

##To Do For Dec 17

• Make slides
• njet weights: data/mc in mumu, emu (g1b) and resulting DY/Z weight in nice plot
• nsearchbin plot with final prediction + any systematics we have
• normalization weight in tight region
• MC closure in backup
• shape variations: HT etc
• 0b tight control region, data vs mc, nsearchbin

1. Method
2. Njet weight systematic
3. Nb=0 normalization
4. shape unc from loose region
5. Final prediction

• Make signal region BG estimate for comparison plots for Zhenbin (https://gist.github.com/zhenbinwu/ce4e9034971dac4236f1)
• Loose HT plots (200, 300, 400, 500 GeV HT cuts)
• Systematics
• Tight N(b) = 0 nSearchBin plot for data and MC
• Apply normalization weight to DY (make new histo file, and run makeplots)
• After Njet weighting get Data/MC shape factors for
• MET
• MT2
• Njet
• Nb
• Nt
• Jet Pt
• Muon Pt
• Ht
• NJet weight uncertainties
• N(b) normalization scale uncertainties

##To Do Thanksgiving

• check data vs MC
• Check MC closure in relaxed cut region
• Implement cutflow functionality
• Create plots of Inclusive DY MC
• Implement lhe HT
• Improve structuring of makePlots
• Implement minituples for systematics
• Implement code to analyse minituples and create systematic outputs
• Fix MC closure
• Implement trigger/lepton efficiencies

##To Do List

plots we want to see

• real b fraction (gen b) of each sample
• harder than it sounds, insufficient gen info in NTuples
• CSV values of faked jets
• Kinematic properties/n-jet in low vs. high CSV range
• Kinematic n-jet in 0,1,2,3 b bins

Ntuple production

• Run over singlemuon data
• Run over TTZ MC

Code work

• Table dump for cutflows
• Update samples.cc with latest sample information
• Make Hongxuan stop using samples2.h
• Top tagger b-faking improvements
• Implement lepton efficiencies
• Implement b-tag scale factors
• Write smaller fitter for the shape systematics

Analysis

• Check photon cleaning around muons
• Is this really the muon cleaning bug?
• check closure, particularly in high Zpt/MET and N(b) = 0 -> N(b) > 0
• data/MC comparison
• raw kinematic quantities (ttbar and DY)
• compare DY->mumu MC to data samples
• ttbar selection using emu selection
• leading/sub-leading jet pt/dR issues
• recalculate acceptance, and efficiency corrections with Spring15 samples
• recalculate gen z-smearing,
• Check closure of b-faking method in loose region

Documentation

• do it