KinZfitter

Using Z mass kinematic constraint(s) to refit lepton momenta in H-to-ZZ*

• Code structure

KinZfitter : the class read the inputs from leptons and fsr photons that form the Higgs Candidate,, do the refitting, and get the refitted results HelperFunction : the class that read lepton/photon pT errors by accessing the pat:Electron, pat:Muon and pat:PFPartticle and also provide the function to help calculate mass4l error (including the covariance matrix)

To include the refit in your analyzer:

0.Check out package

cd $CMSSW_BASE/src

git clone https://github.com/tocheng/KinZfitter.git

cd KinZfitter

git checkout -b from-v1.0 v1.0

cd ../

scram b -j 8

(Tag for Moriond 2016 will be named as v1.x )

In your main analyzer:

0.Add the package into your BuildFile.xml

1.include the head file

#include "KinZfitter/KinZfitter/interface/KinZfitter.h"

2.Declare and then initialize the KinZfitter class when initializing your analyzer i.e., declare and initialize in beginJob() in EDAnalyzer framework KinZfitter *kinZfitter; kinZfitter = new KinZfitter(isData); //(In data, (isData=true). In mc (isData=false))

3.Prepare inputs after Higgs candidate is formed:

leptons: Suppose Lep_Z1_1,Lep_Z1_2, Lep_Z2_1,Lep_Z2_2 are pat:Electron or pat:Muon from Z1 and Z2 decay.

In your analyzer, do:

 vector<reco::Candidate *> selectedLeptons;
 reco::Candidate *cZ1_1 = dynamic_cast<reco::Candidate* >(&Lep_Z1_1);
 reco::Candidate *cZ1_2 = dynamic_cast<reco::Candidate* >(&Lep_Z1_2);
 reco::Candidate *cZ2_1 = dynamic_cast<reco::Candidate* >(&Lep_Z2_1);
 reco::Candidate *cZ2_2 = dynamic_cast<reco::Candidate* >(&Lep_Z2_2);
 selectedLeptons.push_back(cZ1_1);
 selectedLeptons.push_back(cZ1_2);
 selectedLeptons.push_back(cZ2_1);
 selectedLeptons.push_back(cZ2_2);

fsrPhotons : Supporse find pat::PFParticle fsrPhoton which is the fsr photon. To fill the array if the photon is associated to a certain lepton from Z1 or Z2 decay, do something like:

 TLorentzVector p4FSR(fsrPhoton.px(),fsrPhoton.py(),fsrPhoton.pz(),fsrPhoton.energy);

 std::map<unsigned int, TLorentzVector> selectedFsrMap;
 if(associateLeptonZ1_1) selectedFsrMap[0] = p4FSR;
 if(associateLeptonZ1_2) selectedFsrMap[1] = p4FSR;
 if(associateLeptonZ2_1) selectedFsrMap[2] = p4FSR;
 if(associateLeptonZ2_2) selectedFsrMap[3] = p4FSR;

the key of the map stands for the position of leptons in selectedLeptons vector,
and the value is the TLorentzVector of the fsr photon 4-vector.

4.Setup, refit and get the refitted results:

In your analyzer, do

  kinZfitter->Setup(selectedLeptons, selectedFsrMap);
  kinZfitter->KinRefitZ1();
  
  // refit mass4l
  double mass4lREFIT = kinZfitter->GetRefitM4l();
  // four 4-vectors after refitting order by Z1_1,Z1_2,Z2_1,Z2_2
  vector < TLorentzVector > p4 = kinZfitter->GetRefitP4s(); 
  // refitted mass4l error
  double mass4lErrREFIT = kinZfitter->GetRefitM4lErrFullCov();

There is a function called GetRefitM4lErr() which calculates mass4l error after refitting assuming that all lepton momenta are UNcorrelated which is good approximation for reco lepton momenta BUT UNTRUE for refitted lepton momenta. This function will be extended to calculated the mass4l using full covariance matrix soon.

5.Support functions

massZ1REFIT

double massZ1REFIT = kinZfitter->GetRefitMZ1();

massZ1Err

double massZ1Err = kinZfitter->GetMZ1Err();