Update Update Karol Krizka and Michele Petteni

Document Sample
Update Update Karol Krizka and Michele Petteni Powered By Docstoc
					            Update

Karol Krizka and Michele Petteni

       Simon Fraser University


      CERN, 23/06/2009
  Outline




           Variable Definitions
           Possible Backgrounds
           Building Z → nunu Templates
           Building Z → ll/W → lnu Templates




Karol and Michele (Simon Fraser University)   Update   23/06/2009   2 / 28
  Variable Definitons

                                                            →
                                                            −    →
                                                                 −
                                          HT =              PT + E T
                                                 TwoJets


                                              −→
                                              −−                     →
                                                                     −
                                              MHT =                  PT
                                                           TwoJets


                                                  −→ −
                                                  −−    →
                                                  MET = ET




                   Note: For the vector variables, we plot the magnitude.

Karol and Michele (Simon Fraser University)           Update              23/06/2009   3 / 28
  Background Sources



           Electroweak Background
                   Missing ET
                           Z → nunu
                           Unreconstructed Z → mumu and W → munu
                   Extra Jets and Missing ET
                           Z → ee
                           W → enu
           QCD Backgrond




Karol and Michele (Simon Fraser University)    Update              23/06/2009   4 / 28
  Generator Level Cuts



   Z → nunu has a generator level filter applied, so is inconsistent with the
   other samples.
   Filter:
        Require at least one jet with
                   PT > 30 GeV
                   Eta < 5
           Based on Cone4TruthJets




Karol and Michele (Simon Fraser University)   Update             23/06/2009   5 / 28
  Generator Level Cuts - Solution




   The following actions were taken to resolve the inconsistency:
           TruthJetFilter copied and applied to the dataset from
           CSC.007712.AlpgenJimmyZnunuNp2 pt20 filt1jet.py
           Jets matched to electrons comming from Z/W were removed from
           the jet collection




Karol and Michele (Simon Fraser University)   Update                23/06/2009   6 / 28
  Building Z → nunu Templates



   We cannot directly build a data-driven template for Z → nunu, because
   neutrinos are not reconstructed. But...

           ET = Two missed neutrinos +Other Missed Particles + Cracks
           ET =                   Missed Z     +Other Missed Particles + Cracks
           ET =          Two missed leptons    +Other Missed Particles + Cracks

   We can pretend that we missed two leptons by removing them from the
   events and recalculating all of the variables. This works, because ET
   distribution of the Z should be the same for all channels.




Karol and Michele (Simon Fraser University)   Update                   23/06/2009   7 / 28
  Truth Boson ET Distributions




                                              Based on Monte Carlo
Karol and Michele (Simon Fraser University)           Update         23/06/2009   8 / 28
  Algortihm Flow




   Run over different number of partons, and combined based on cross
   section.


Karol and Michele (Simon Fraser University)   Update         23/06/2009   9 / 28
  Cross Sections



                                      σ =σ×           TruthJetFilter   ×   My Cuts


       Process                  Np0            Np1      Np2         Np3         Np4     Np5      Total
       Z → nunu                0.000          1.318    18.291      8.461       2.042   0.305     30.42
       Z → mumu                0.003          0.169    1.794       0.821       0.189   0.031     3.01
       Z → ee                  0.000          0.087    0.989       0.454       0.112   0.017     1.66



   σ taken from:
   https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TopReferences10TeV



Karol and Michele (Simon Fraser University)               Update                         23/06/2009   10 / 28
  My Cuts




           First Leading Jet ET > 30GeV
           Second Leading Jet ET > 30GeV
           Third Leading Jet ET < 30GeV
           Missing ET > 100GeV
           At least two reconstructed leptons (except Z → nunu)




Karol and Michele (Simon Fraser University)   Update              23/06/2009   11 / 28
  Missing ET




Karol and Michele (Simon Fraser University)   Update   23/06/2009   12 / 28
  Missing HT




Karol and Michele (Simon Fraser University)   Update   23/06/2009   13 / 28
  HT




Karol and Michele (Simon Fraser University)   Update   23/06/2009   14 / 28
  Building Z → ll / W → lnu Templates



   Further assumptions:
           There is no difference between reconstructed and non-reconstructed
           Z’s.
           There is no difference between non-reconstructed W’s and
           reconstructed Z’s.
   If these assumptions are true, we can combine all of the electroweak
   background into one template.




Karol and Michele (Simon Fraser University)   Update             23/06/2009   15 / 28
  My Cuts




           First Leading Jet ET > 30GeV
           Second Leading Jet ET > 30GeV
           Third Leading Jet ET < 30GeV
           Missing ET > 100GeV
           No Reconstructed Leptons




Karol and Michele (Simon Fraser University)   Update   23/06/2009   16 / 28
  Background Cross Sections




       Process                  Np0            Np1      Np2      Np3     Np4     Np5      Total
       Z → nunu                0.000          1.318   18.291    8.461   2.042   0.305     30.42
       Z → mumu                0.000          0.007    0.117    0.046   0.012   0.001     0.18
       Z → ee                  0.000          0.000    0.000    0.000   0.000   0.000     0.00
       W → munu                0.003          0.546    6.639    2.941   0.719   0.106     10.95
       W → enu                 0.004          0.761    2.406    0.989   0.220   0.037     4.42




Karol and Michele (Simon Fraser University)            Update                     23/06/2009   17 / 28
  Missing ET




Karol and Michele (Simon Fraser University)   Update   23/06/2009   18 / 28
  Missing HT




Karol and Michele (Simon Fraser University)   Update   23/06/2009   19 / 28
  HT




Karol and Michele (Simon Fraser University)   Update   23/06/2009   20 / 28
  Electron-Jet Contamination
   It is possible that the electron in W → enu is detected as a jet, but not
   reconstructed as an electron. This would drive missing ET down.
   Solution:
         Cut on the EM fraction of the two leading jets such that
         0.1 < EM Fraction < 0.9




Karol and Michele (Simon Fraser University)   Update             23/06/2009   21 / 28
  W → enu Cut On EM Fraction, No MET Cut




Karol and Michele (Simon Fraser University)   Update   23/06/2009   22 / 28
  W → enu Cut On EM Fraction




Karol and Michele (Simon Fraser University)   Update   23/06/2009   23 / 28
  Detector Problems




   It is possible that the difference between the reconstructed and
   non-reconstructed plots could be due to kinematics that kept the leptons
   from being reconstructed. For example, the muon chambers end at Eta of
   about 2.6. Test:
           Compare the truth ET of the reconstructed and non-reconstructed Z’s
           Define a lepton as reconstructed if it lies within eta of 2.6




Karol and Michele (Simon Fraser University)   Update                 23/06/2009   24 / 28
  Truth Z ET Distributions




                                              Based on Monte Carlo
Karol and Michele (Simon Fraser University)           Update         23/06/2009   25 / 28
  Truth Z PT Distributions




                                              Based on Monte Carlo
Karol and Michele (Simon Fraser University)           Update         23/06/2009   26 / 28
  Reconstructed Z → mumu Missing ET




Karol and Michele (Simon Fraser University)   Update   23/06/2009   27 / 28
  Detector Problems




           The ET of the Z depends on eta and exhibits the same pattern as in
           the reconstructed vs non-reconstructed missing ET plot
           The PT of the Z depends on eta, but not as much as ET
           We are currently modifying the missing ET of the event based on the
           leptons’ ET . Maybe we should switch to PT ?




Karol and Michele (Simon Fraser University)   Update             23/06/2009   28 / 28
  Z → mumu Missing ET , Modified With Mu’s PT




Karol and Michele (Simon Fraser University)   Update   23/06/2009   29 / 28

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:8
posted:11/21/2011
language:English
pages:29