Purpose: Produce a more flexible ntuple

  • To produce a very flexible ntuple, which can be used in root environment for almost all the jobs.
    • produce a preselected ntuple with loose cuts or even no cuts.
    • to be able to reconstruct events in ttbar.cxx, too, for some predefined convenient ntuple leaves.
    • be able to used for analysis of di-leptonic ttbar channel
    • fill the joboption setting in ntuple. Set it only once in initialization(first event).
    • readout a complete truth decay chain in a independent code and used by all other needs.

Strategy

  • In DoElectron, DoMuon and DoBJet
    • Do only pre-selection with loose cuts and fill the ntuple
    • Each of them can be used standalone for specific analysis in root env, for example, zee, zuu, zbb etc..
  • in ttbar
    • Take the preselected container, use needed cuts to choose objects.
    • Do event reconstruction.
    • For other analysis other than ttbar, ttbar can be switched off
  • divide long function to be as short as possible, and refer to them in a main program.
  • For truth infomation
    • readout only in ttbar and compose a complete chain for reference

optimization

  • In DoBJet
    • re-valuate the jet calibration to quarks ( should include gluons which decay to jets, include all radiated quarks which decay to jets )
    • or using the official jet calibration method.

bugs

  • jet precalibration
    • for jet calibrate to MC jet, the scale factor is computed from reco jet to truth jet (both no muon included), reco jet is rescaled to truth jet with this factor, and add muon finally.
    • for jet calibrate to quarks, the scale factor is computed from reco jet to truth quark(only jet no muon included). That reco jet is rescaled to truth quark is a correct and rough procedure, but we rescale (reco jet + muon) with this factor, which is not reasonable. We can fix it use the first procedure, or recompute the factors.
  • atlfast jet calibration to quarks
  • similar bug as above for reading truth info of electron and muon: asked the lepton status to be 2
    • for AcerMC events: need to change in DoElectron.cxx and DoMuon.cxx, read out truth lepton in hard process whose mother w is status is 3
    • for mcnlo events: we need a different code for reading truth info
  • In DoBJet.cxx
    • B jet and light jet mis-calibration should done before selection cuts.
  • atlfast b random tag
    • this bug didn't affect analysis result. Filling weightTag Ntuple and using weightTag to tag b, program retrieve random weight twice.
  • read trth
    • in ttbar.cxx, in t->t->t->t->w->w->w, try to read out the last t and the first w. the last t is correctly read out, but in fact read out the second w.
    • don't know the effect
    • in w->tau(neu)->w neu(neu), in finding w, the second w overwritten the first one. in fact we need the first one.

other issues should be considered

Edit | Attach | Watch | Print version | History: r13 < r12 < r11 < r10 < r9 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r13 - 2008-09-16 - ChengguangZhu
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback