Skip to content

Latest commit

 

History

History

vertex_investigation

README.md

Vertex Investigation for nanoAODs

This repo contains the script to perform studies on Hgg vertex, in order to understand if it is possible to use vertex 0th instead of the one we choose with a BDT. The reason for this is that vertex 0th is the only one already present in the centrally produced nanoAODs. Using this branch of flashgg we produce two samples of simulated events for the four main Higgs production mechanisms (ggH, VBF, VH, ttH), one with the currently used vertex and the other with vertex 0th. To run it, after checking out, simply run:

./Validation/test/run_vertex_investigation.sh
./Validation/test/run_vertex_investigation_vtx0.sh

The scripts in this repo perform studies on these two samples and compare them.

These are the main studies performed:

  • Reconstruction efficiency: compute fraction of vertexes are reconstructed closer than 10 mm from the "true" one; this is done as a function of pt (draw_efficiencies_pt.py) and number of vertexes (draw_efficiencies_nvtx.py); resulting plots can be seen here;
  • Fit $\sigma_M$ in subdetector categories, performed in two ways: using zfit (fit_sigma_m.py, even if goodness of fit is not performed) and using RooFit (fit_sigma_m_roofit.py); the categories chosen for the fit are (for each gamma pair) EBEB, EBEE, EEEE; main plots available [here];(https://gallim.web.cern.ch/gallim/plots/Hgg/VertexInvestigation/mass_fit_subdetector_categories/mass_fit_roofit/)
  • Fit $\sigma_M$ in categories of $\frac{\sigma_{M}}{M}$: categories are designed with an algorithm that minimizes the difference between number of events in each bin; fit_sigma_m_smom_cat.py performs the fit and dumps a pickle file (sigma_m_final_plots_specs.pkl) which then works as an input for fit_sigma_m_smom_final_plots.py to produce the final plot (see here).

All the scripts mentioned above can be run simply by typing (taking as an example draw_efficiences_pt.py):

python draw_efficiences.py --output_dir path_to_output_dir --channel ggH/VBF/VH/ttH

in some cases, it might be necessary to also specify --v0-input-dir and --vcustom-input-dir to specify where the ntuples for the 0th vertex and custom vertex (respectively) are stored.