Run 3 R(K) Trigger Efficiency Studies

This repository calculates the trigger efficiencies and Scale Factors (SFs) for the Run 3 R(K) analysis using an orthogonal dataset method (DoubleMuon).

Setting Up

Building Environment (First-Time Setup)

cmsrel CMSSW_14_0_6
cd CMSSW_14_0_6/src
cmsenv
git clone git@github.com:DiElectronX/r3k-trigeffs.git
cd $CMSSW_BASE/src/r3k-trigeffs

Loading Environment (When Logging In)

cd .../r3k-trigeffs
cmsenv

Procedure Overview

We utilize the 2022 DoubleMuon dataset (NanoAOD) as an orthogonal proxy to measure the efficiency of the di-electron triggers. The workflow consists of three main stages: Skimming, Histogramming, and Plotting/SF Calculation.

We employ a mixture of L1+HLT trigger paths. Scale Factors are derived by comparing Data (fitted J/psi yields) vs MC (cut-and-count weighted by luminosity fractions).

Note: Ensure you inspect all .yml config files before running to confirm dataset paths, cuts, and trigger lists are correct for your specific study.

---

Step 1: Skimming

This step filters the NanoAOD events. It:

1. Uses a JSON filter to select events where the di-electron triggers were active.
2. Selects events firing a subset of DoubleMuon triggers that are mostly orthogonal to the electron component.
3. Applies rough cuts to emulate the di-electron phase space (>=2 opp. sign electrons, small dR/dz, pT/eta acceptance).

Configuration: eff_skim_cfg.yml

Command:

python3 skim_nanoaod.py -c eff_skim_cfg.yml

---

Step 2: Histogramming

This step bins events for the numerator and denominator of the efficiency calculations.

• Data: Iterates through events, checks trigger flags, and fills num/denom histograms based on the subleading electron . This is performed for each trigger path in the mixture and the logical OR.
• MC: Similarly fills histograms but applies MC event weights based on FONLL corrections to the B meson spectrum to correct for mismodeling.

Configuration: eff_hist_cfg_v2.yml

Commands:

# Process Data
python3 make_data_efficiency_hists_v2.py -c eff_hist_cfg_v2.yml

# Process MC (with FONLL weights)
python3 make_mc_efficiency_hists_v2.py -c eff_hist_cfg_v2.yml

---

Step 3: Plotting and Scale Factors

This step calculates efficiencies, plots them, and exports the Scale Factors (SFs).

• Data: Loads histograms bin-by-bin and fits a J/psi peak to extract signal yields, which are used to calculate efficiency per bin.
• MC: Performs a cut-and-count efficiency calculation for each trigger path. A weighted sum of path-by-path efficiencies is calculated based on the exclusive luminosity fractions (defined as where a specific path is the lowest unprescaled, operational trigger).
• Output: Plots of efficiencies, Data/MC ratios, and a JSON file containing the efficiencies and SFs.

Configuration: eff_plot_cfg_v2.yml

Command:

python3 make_mc_efficiency_plots_v2.py -c eff_plot_cfg_v2.yml

---

Analysis Application

For the general analysis using these corrections:

1. Allocation: Allocate MC samples to specific L1+HLT paths according to the exclusive luminosity fractions used in Step 3.
2. Filtering: Filter MC events based on whether they pass their allocated trigger path.
3. Correction: For passing events, apply the Scale Factor (SF) derived for the whole trigger mixture.

Note: This mixture approach was adopted because statistics were insufficient to derive robust SFs binned in multiple dimensions (pT, dR) for individual trigger paths.