Skip to content

Latest commit

 

History

History
117 lines (78 loc) · 3.74 KB

README.md

File metadata and controls

117 lines (78 loc) · 3.74 KB
Raw

LimitComputation_MassSpectrum

This repositery is dedicated to HSCP analysis, for the limits computation with the mass spectrum approach.

Setup working area

export SCRAM_ARCH=slc7_amd64_gcc700
cmsrel CMSSW_11_3_4
cd CMSSW_11_3_4/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 https://github.com/DenkMybu/LimitComputation_MassSpectrum.git LimitComputation_MassSpectrum

Install the Combine packages and setup

git clone https://github.com/cms-analysis/HiggsAnalysis-CombinedLimit.git HiggsAnalysis/CombinedLimit
cd HiggsAnalysis/CombinedLimit

Update to a recommended tag - currently the recommended tag is v9.1.0 More information on: https://cms-analysis.github.io/HiggsAnalysis-CombinedLimit/

cd $CMSSW_BASE/src/HiggsAnalysis/CombinedLimit
git fetch origin
git checkout v9.1.0
scramv1 b clean; scramv1 b # always make a clean build

Then install Combine Harvester package. More information on: https://cms-analysis.github.io/CombineHarvester/

cd CMSSW_11_3_4/src
cmsenv
git clone https://github.com/cms-analysis/CombineHarvester.git CombineHarvester
git checkout v2.0.0
scram b

Create HSCPLimit repository:

mkdir HSCPLimit
cd HSCPLimit

You are now ready to create the datacards in the LimitComputation_MassSpectrum directory using the available scripts.

Import the cross-sections

The cross-section are already loaded in the directory xsec. More information on: https://github.com/fuenfundachtzig/xsec

Create the datacards

This part aims to create the datacards for each signal hypotheses.

Use the script create_datacards.py:

python create_datacards.py

Set the variable regionSignal to 'SR1', 'SR2' or 'SR3' depending on what you want. Set the variables pathSignal and pathPred to open signal and predict mass distributions. The datacards are created for a whole bunch of different signal hypotheses and use the different predicted mass shapes due to systematics. Set the output directory with the variable outDataCardsDir. Setting the variable systSignal one will produce systematics budget for a given signal hypothesis, as a function of target masses (so systematics budget within the mass windows).

The bias correction parameters are currently hard-coded for each signal regions in both year. Be sure to update the values obtained with the package massSpectrum_bckgPrediction, especially using the macroMass.py script (see: https://github.com/dapparu/massSpectrum_bckgPrediction).

Run Combine on the datacards

This part gives the way to run Combine limits and Combine significance on the previously produced datacards.

Use the script run_datacards.py:

python run_datacards.py

Set the input_dir and the tree_dir variables. The input directory of this code is the output directory of the previous stage.

By default, the code run the AsymptoticLimits and Significance methods, in parallel.

All the results are saved in the tree_dir directory.

Produce the limits plots

One uses the limit_plots.py script to produce limits plots:

python limit_plots.py

Set the labelSignal variable to extract limits on a given signal hypothesis. The supported hypothesis are the commented ones.

The results are saved on the limit_plots_dir directory.

Significance computation

Work in progress...

Signal injection tests

Work in progress...

Mass shape analysis

Work in progress...