November 2007
At first I started off by calculating the missing transverse energy corresponding to lepton neutrino. It is easy to get the MET through some modules, but to calculate total missing energy is not easy rather hard-working. Therefore I have chosen the fully hadronic channel because it is more instructive at the moment. After finishing fully hadronic channel the semi-leptonic channel will be taken into account. I suppose the study of fully hadronic channel could be helpful to analyze the semi-leptonic channel because it has not only leptonic channel but also half hadronic channel. Jeewon also has started studying fully hadronic channel. However we are following slightly different ways. I focused on the event selection cuts and she centred on QCD background.13.11.2007
Fully hadronic channel
- Some features :
- Large QCD backgrounds
- Need not to take into account the missing energy (neutrinos)
- Largiest branching ratio ~ 46%
- Two leading jets are considered as b-jets
- b-tagging are not performed in this analysis but it will be taken into account later
Framework and samples
- Framework
- CMSSW_1_5_2
- SusyAnalyzer
V01-01-00
- ROOT
- Samples
- tt0j_mT_70_alpgen ~ tt4j_mT_70_alpgen
- DBS data discovery page
- QCD backgrounds are contained at the generation stage.
- DBS data discovery page
- tt0j_mT_70_alpgen ~ tt4j_mT_70_alpgen
Preliminary
Jet efficiency
- Tried to make efficiency plots by varying the number of jets in events
- Tried to see the effect to the efficiency by adjusting the parameters corresponding to jest in analysis.cfg
- Event rejection decision : rej_BadHardJet = TRUE ro FALSE
- Primary and final accpetance pt cut, 20 GeV and 30 GeV, respectively
- Also eta cuts for primary and final acceptance : Max = 5
- Extrapolation error adjustment(in cm) :
- reco_jetD0ErrorThresh = 0.05
- reco_jetDzErrorThresh = 0.10
- reco_bjetD0ErrorThresh = 0.5
- reco_bjetDzErrorThresh = 0.20
Event selection
Sum of non-leading jests
- Non-leading jet total transverse energy obtained removing the two most energetic jets
- Sum_3 >= 148 GeV
Verifying jet-parton-matching
- The six partons in fully hadronic channel are matched to six reconstruced jets by picking the matching which minimises the sum of the angular seperation between reconstructed jet and matched parton.
CMS Notes- good jet-parton-matching : Each of the six partons and jets differ only by 15(deg) and the jet-reconstructed tops also differ only by 15(deg) from their corresponding parton-level direction.
- half-good jet-parton-matching : Three of the partons and jets forming one top differ only by 15(deg) and this jet-reconstructed top also differ only by 15(deg) from his corresponding parton-level direction.
- bad jet-parton-matching : Everthing else.
- SusyAnalyzer has own MatchObject class, so I tried to verify how it is efficient by using this method.
- As results, the value of 15(deg) is somewhat arbitrary but all matching objects are good.
- i.e. total six jets and partons, so 6 * 15 = 90(deg). the distribution for good jet-parton-mathing shows well below 90(deg)
The other event selection cuts
Centrality
- Centrality C >= 0.68
- the fraction of the hard scatter energy going in the transverse plane.
Aplanarity
- Aplanarity A >= 0.024
- A = 3/2 * Q : Q being the smallest of the three normalized eigenvalues of the sphericity tensor
- Choose 3rd eigenvalues in TMatrixDEigen Class
- Sphericity tensor
- Calculating Eigen values of Sphericity tensor
- How to use the linear algebra packages(LAPACK) in ROOT
- Matrix references : http://root.cern.ch/root/html/MATRIX_Index.html
- Matrix references : http://root.cern.ch/root/html/MATRIX_Index.html
- TMatrixT
, TMatrixDEigen
- How to use the linear algebra packages(LAPACK) in ROOT
Computing
OldWebs/OldWebs/DAQ Web
Create New Topic
Index
Search
Changes
Notifications
Statistics
Preferences |
|
Copyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors. Ideas, requests, problems regarding KoreaCmsWiki? Send feedback