########################################################## #########################################################

IMPORTANT: To run the updated analysis code, run runFullTauIDmeasurement.sh

########################################################## #########################################################

This is collection of RooT macros used to perform measurements of the tauId efficiency with W*->tau+v events.

PlotTrigger.C - macro to compute trigger turn on curves for MET trigger. The macro produces as an output RooT file trigger_eff.root with functions describing efficiency of the MET trigger in dependence of offline MET and MHT variable. The newly created file should be placed in the directory $CMSSW_BASE/src/src/DesyTauAnalyses/NTupleMaker/data and specified in the configuration file

ComputeFakeRate.C - computes fake rate as a function of tau Pt. One can analyze either W(->mu+v)+Jets or dijet samples. The macro produces as an output the RooT file ${dataset}_fakeRate${tauIsolation}.root where ${dataset} is either SingleMuon_Run2016 or JetHT_Run2016, and ${tauIsolation} is tauId working point : LooseIso, MediumIso, TightIso, LooseMvaIso, MediumMvaIso or TightMvaIso.

PlotWTauNuBkgd.C - computes fraction of electroweak events in the anti-isolated tau region of the W*->tau+v selection.

PlotFakeRate.C - plot fake rates and compute combined fake rate for measurements in W(->mu+v)+Jets or dijet samples. The macro prints out combined fake for different fake tau pT bins. The numbers are currently cut and pasted in the AnalysisNTupleProducer.cc macro. TODO : save results in the form of TH1D or TGraphAssymErrors to the RooT file, which can be used by the analysis macro to apply fake rate to data events in the anti-isolated region.

PlotWMuNu.C - plots mT(mu,MET) (or any other distribution) in the selected W*->mu+v sample and creates datacards for statistical inference.

PlotWTauNu.C - plots mT(tau,MET) (or any other distribution) in the selected sample of W*->tau+v events and creates datacards for statistical inference.

PlotWMuNuPostfit.C - plots postfit distribution of mT in the W*->mu+v sample given results of the fit with combine tool

PlotWTauNuPostfit.C - plots postfit distribution of mT in the W*->tau+v sample given results of the fit with combine tool.

All ntuples are located in /nfs/dust/cms/user/rasp/Run/Run2016/TauID_2016

The macro AnalysisNTupleProducer is executed with two arguments, configuration file and filelist.

Examples of configuration files are

• analysisNTupleProducer.conf (to run on data)

• analysisNTupleProducer_MC.conf (to run on MC)

Examples of filelists :

MET_Run2016G

WToTauNu_M-200_13TeV-pythia8

WToMuNu_M-200_13TeV-pythia8

Examples of running macro

AnalysisNTupleProducer analysisNTupleProducer.conf MET_Run2016G

AnalysisNTupleProducer analysisNTupleProducer_MC.conf WToMuNu_M-200_13TeV-pythia8

Useful scripts :

qsub.sh - submits job to batch system. This script requires three arguments :

1. code name

2. config file

3. filelist

Example :

qsub.sh AnalysisNTupleProducer analysisNTupleProducer.conf MET_Run2016G

qsub_seq.sh - submits several jobs per filelist. The script requires four arguments :

1. code name

2. config file

3. filelist

4. files per job

The script creates directory named $filelis_files. The directory will contain all RooT and output files created by separate jobs. Once all jobs are finished, all RooT files can be merged by executing script hadd.sh

hadd.sh $filelist.

The directory datacards contains all datacards for statistical inference and results of the fits.

The datacards from two regions - signal (W*->tau+v) and control region (W*->mu+v) are combined using script CombineCards.sh. This will produce datacards for various tau ID working points

tauId_LooseMvaIso.txt

tauId_MediumMvaIso.txt

tauId_TightMvaIso.txt

tauId_LooseIso.txt

tauId_MediumIso.txt

tauId_TightIso.txt

To run computation of the tauId, one needs to execute script

Fitting.csh

and pass as an argument tau isolation working point (LooseMvaIso, MediumMvaIso, TightMvaIso, LooseIso, MediumIso or TightIso)

Example :

./Fitting.sh MediumIso

This will create mlfit_tauId_MediumIso.root. This file is then used to plot posfit distributions of fitted variables (mT(mu,MET) in the W*->mu+v control region and mT(tau,MET) in the W*->tau+v signal region).