cmsdiphoton/analysiscode

AnalysisCode

** Analysis ** This directory contains the necessary code to make the analysis histograms from ggNtuples.

• AllieCode/ana.C defines what files will be used.
• AllieCode/BatchMode has the files you need to run the analysis jobs over condor at the LPC. Note that the LPC recently refactored their condor farm, so this might not work out of the box. First run setupCondor.sh, then make sure anaMASTER.C is pointing to the right datasets and file locations, and then submit the jobs with PythonSubmit.py. There is also a script to clean up all of the files after the jobs run.
• AllieCode/ggEventTree.h defines the variables in the ggNtuples that will be used by the analyzer
• AllieCode/SusyEventAnalyzer.cc is the analysis code that sorts the events into signal and control regions and makes all the histograms.
• Diempt.C will take the input gg and ff or ee histograms of diempt and calculate the diempt ratios which can be used to reweight the MET distributions. This macro also makes a nice plot of the diempt ratio and calculates the appropriate errors.
• FinalAnalysis.C takes the output of the event loop and calculates the final background estimates and errors and makes several plots of the final quantities.
• GGM_Combination/finishHists_*C calculates the uncertainties that we implemented for the GGM combination paper.
• lester_mt2_bisect.h is the code to calculate MT2.
• ratioMethod.C performs the fit to the gg/ff ratio.

** Limits ** Code to make limit plots from the datacards.

• MakeCountingFiles.ipynb is a jupyter notebook for creation of the datacards and scripts needed to run combine
• MakeLimitPlots.ipynb is a jupyter notebook for creation of limit plots from the counting files created by combine
• mlfit.ipynb and .py creates diagnostic plots of the limit calculation
• PlotsSMSFiles is a directory that contains all the files you need to edit to use the PlotsSMS package to make nice looking limit plots.

** NtupleCode ** Code to manipulate the ggNtuples.

• Condor_skimNtuples contains example code to submit condor jobs on the LPC to skim ggNtuples and only select interesting events.
• sortSignalByMass.py will take the events in a ggNtuple and sort it by squark/gluino vs neutralino mass.

** Utils ** This directory contains random useful pieces of code

• FinalPlot.C uses the output of FinalAnalysis.C to produce the final pretty stack plot.
• GridDiemptContamination.C will calculate the signal contamination in the gg diempt distribution from each of the signal mass bins.
• PrintBinContents.C is useful for debugging - it will print the contents of a histogram or its errors.
• accXeff.C can be used to make the acceptance x efficiency plots for the T5gg or T6gg grids.