massSpectrum_bckgPrediction

This repositery is dedicated to HSCP analysis, for the background estimate of the mass spectrum.

Setup working area

export SCRAM_ARCH=slc7_amd64_gcc700
cmsrel CMSSW_10_6_27
cd CMSSW_10_6_27/src/
cmsenv

For the following step you should have a ssh key associated to your GitHub account. For more information, see connecting-to-github-with-ssh-key.

git clone -b master [email protected]:dapparu/massSpectrum_bckgPrediction.git massSpectrum_bckgPrediction 

Input file path

The path of input file is given here:

• Unblinded production: /opt/sbg/cms/ui3_data1/dapparu/HSCP/Production/crab_Analysis_SingleMuon_Run2017_CodeVUnB_v1_v1.root and /opt/sbg/cms/ui3_data1/dapparu/HSCP/Production/crab_Analysis_SingleMuon_Run2018_CodeVUnB_v1_v1.root. You will also find files for the different eras.
• Blinded production: /opt/sbg/cms/ui3_data1/dapparu/HSCP/Production/crab_Analysis_SingleMuon_Run2017_CodeV73p3_v4.root and /opt/sbg/cms/ui3_data1/dapparu/HSCP/Production/crab_Analysis_SingleMuon_Run2018_CodeV73p3_v4.root

Run the background estimate code

This part concerns the run of the background estimate method.

python launchBckgEstimateOnAll.py

The following script allows to run all the configurations associated to each systematic automatically. You should use a screen because it may take several days (runs each region for both 2017 and 2018) :

python test_launch_all.py

You can change on what you want to run directly in launchBckgEstimateOnAll.py and step2_backgroundPrediction.C files.

In the first file (launchBckgEstimateOnAll.py):

• You can set on which datasets you want to run, giving the path to the root file with all the needed histograms, in the confidatasetListg array.
• The config array is used to indicate which kind of estimates are ran: nominal or the different systematics.
• The nPevariable set the number of pseudo-experiments done during the background estimate.
• The odir array gives the directory where you can find fast produced plots.

The code runs on 25 cores in parallel (in local) and it can be changed at line 798 of the file Regions.h

In the second file (step2_backgroundPrediction.C):

• You can set which regions you want for estimation. Default: only SR1, SR2 and SR3.

After you ran the code, a new file is created with the histograms corresponding to the background estimate in the wanted regions, and labels are set to correspond to the different systematics cases.

Important Normalisation problem in CR In the case you see normalisation problems (especially in control regions), be sure to apply the Ih > C cut at the preselection stage.

Run the mass spectrum plotter code

This part concerns the run of mass spectrum plotter code, with nice style. The path where all the scripts are is :

/opt/sbg/cms/ui6_data1/rhaeberl/CMSSW_10_6_30/src/massSpectrum_bckgPrediction

The code runs with the command:

python2.7 macroMass.py --ifile /opt/sbg/cms/ui3_data1/dapparu/HSCP/Production/crab_Analysis_SingleMuon_Run2017_CodeVUnB_v1_v1_cutIndex3_rebinEta4_rebinIh4_rebinP2_rebinMass1_nPE200_test_v1.root --ofile test1 --region 999ias100 --odir test

You can also use the following python script to run the plotting macro on all chosen regions :

python ShowPreds.py

Concerniing the options:

• option --ifile is for the input file; obtained at the previous step.
• option --ofile is for the output file label.
• option --region is for the region on which one wants to run. The possible values are: 50ias60, 60ias70, 70ias80, 80ias90, 50ias90, 90ias100, 99ias100 and 999ias100.
• option --odir is for the output directory.

The year on which runs is set directly in macroMass.py at line 204.