Diapositiva 1

1
A (very) preliminary study of channel pp-gth-gtZZ-gt
4mu via gg fusion with CMSSW
Alessandro Giacobbe Cristina Botta Daniele Trocino
Relatrice Chiara Mariotti
2
Signal
gg ? h ? ZZ? 4mu

Leading Order Feynman Diagrams
Torino, 27/03/2007
3
Backgrounds

• ZZ( ) ? 4mu

Leading Order Feynman Diagrams

• Zbb ? 4mu
• tt ? 4mu

Torino, 27/03/2007
4
Pythia MC Generation of events

• gg ? h ? ZZ ? 4mu
• ZZ ? 4mu
• Zbb ? 4mu
• tt ? 4mu

(more events for a better study of the signal)
(as we still dont have more events generated by
Comphep)
We work with the expected Luminosity of the first
2 years
Torino, 27/03/2007
5
CMSSW what we need to know

• The overall software, called CMSSW, is build
  around a Framework, an Event Data Model (EDM) and
  services need by the modules that process data
• Only one executable cmsRun
• Many plug-in modules run algorithms the same
  for detector and MC data

.CFG
MODULE

• used to configure cmsRun at run time
• tells which data to use
• modules to run (all loaded at beginning of the
  job)
• parameters to set

• component of CMSSW code
• can be plugged into cmsRun
• it is called for every Event according to the
  path statement
• 6 differet types (Source,EDProducer, EDFilter,
  EDAnalyzer, EDLooper, OutpuModule)
• configured through .cfg using modul-specific
  ParameterSets

Torino, 27/03/2007
6
CMSSW how it works

• Events are processed by passing the event
  through a sequence of modules
• the module can get data from the Event and put
  data back to the Event (not every module)
• each module can talk to the others only
  through the Event

Torino, 27/03/2007
7
The FourMuFilter
A Filter is a module that we used to apply cuts
in generation.
To be accepted the event must have in the final
state at least 4 mu, 2 positive and 2 negative,
with Pt gt 3 GeV, ? lt 2.5.
Warning
If the Event doesnt satisfy the condition it
wont be automatically rejected. To have in
output only filtered events you need to specify
it in the PooloutModule.
Torino, 27/03/2007
8
The FourMuFilter (2)

• gg ? h ? ZZ ? 4mu
• ZZ( )? 4mu
• Zbb ? 4mu
• tt ? 4mu

GeV

• In the next days our filter will be send to
  Filip Moortgat in order to include it in the next
  MC production

Torino, 27/03/2007
9
Pythia kinematic cuts (1)
Apart from standard settings that we include in
the list of pythia cards ( such as the choice of
the partonic functions, or the swich on of parton
shower and QED bremsstrahlung), we also introduce
the following Kinematics cuts (that behave like
a pythia inner filter)

• CKIN(45) - CKIN(48) (D 12., -1, 12., -1)
• range of allowed mass values of the two secondary
  resonances
• produced in a 2-gt1-gt2 process like gg-gth-gtZZ
• CKIN(41) - CKIN(44) (D 12., -1, 12., -1)
• range of allowed mass values of the primary
  resonances

Torino, 27/03/2007
10
Pythia kinematic cuts (3)
Cuts set at 5.,150.,5.,150.
Default Kinematics Cuts
The situation is the same. We chosed to let this
setting as it seems to be the standard way.
GeV
Torino, 27/03/2007
11
Pythia kinematic cuts (2)
Cuts set at 5.,150.,5.,150.
Default Kinematics Cuts
Useless generated events. We will surely cut this
peack coming from µ generated by virtual photons.
GeV
Torino, 27/03/2007
12
Analysis of the generated variables the
FourMuAnalyzer

• Invariant Mass
• all possible muons pairs
• muons pairs coming from Z ? lower values of
• (in this way we selected candidate Z)
• ZZ pairs

• Transverse Momentum
• pt of all muons
• max, second, third, min pt of muons
• pt of all Z

• Angular Variables
• pseudorapidity of muons and Z
• theta between two muons with same charge
• theta between two muons from Z
• Theta between two candidate Z

Torino, 27/03/2007
13
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Transverse Momentum
Torino, 27/03/2007
14
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Torino, 27/03/2007
15
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Torino, 27/03/2007
16
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Invariant Mass
Torino, 27/03/2007
17
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Torino, 27/03/2007
18
Plots Normalized and Weighted
Generated Variables of Signal (
) and Background
Angular Variables
Torino, 27/03/2007
19
Plots Normalized and Weighted
Torino, 27/03/2007
20
How well do we reconstruct Zs? The FourMuTruth
Z candidates ? µµ- couples nearest to Z mass
To measure how much this method works, we define
two efficiencies
Pythia provides history of each generated event
(production/decay chain) ? its possible to
verify how often µ-couples really come from a
generated Z
We get
for both h ? ZZ ? 4µ and ZZ ? 4µ processes
Main inefficiency cause photons emission in
final state
Torino, 19/03/2007
21
How well do we reconstruct Zs? The FourMuTruth
Pythia also provides information about all
generated particles (E, p, ? ) ? its possible
to access true data (i.e. generated with MC)
and compare them to reconstructed data (i.e.
from selected µ-couples)
Torino, 19/03/2007
22
Higgs produced via VV Fusion

• VV ? h ? ZZ ? 4mu

• Aims
• compare Pythia-Phantom for VV-fusion
• (attempt to) separate VV-fusion signal
• from gg-fusion one

Torino, 19/03/2007
23
Weighted plots expected events
Generated variables of signal gg-fusion, signal
VV-fusion, background

Torino, 19/03/2007
24
Kinematics differences
Generated variables of signal gg-fusion, signal
VV-fusion, background
Normalized plots

Torino, 19/03/2007
25
Kinematics differences
Generated variables of signal gg-fusion, signal
VV-fusion, background
Normalized plots

Torino, 19/03/2007
26
Kinematics differences
Generated variables of signal gg-fusion, signal
VV-fusion, background
Normalized plots

Torino, 19/03/2007
27
What to do
First of all EXAMS !!!!!!!!
Apply cuts
muons Pt muons pairs invariant mass (Z cand.
mass) find new kinematics variables (muons
collinearity, )

Do the same study for final state
Do the complete simulation- Reconstruction
SIMU-DIGI-RECO
Torino, 19/03/2007