Pile-up profile

Produce pile-up profile from data using CMS official tools

Source: TWiki

Example

pileupCalc.py -i MyAnalysisJSON.txt --inputLumiJSON pileup_latest.txt \
    --calcMode true --minBiasXsec 69200 --maxPileupBin 100 \
    --numPileupBins 100 MyDataPileupHistogram.root

where

• MyAnalysisJSON.txt is returned by the CRAB job
• pileup_latest is the input JSON file (2016, 2017, 2018)
• 69200 corresponds to the latest estimation of the minimum-bias cross section in nanobarns.
• 100 corresponds to the number of bins for the output histogram: TH1D("pileup","pileup",0,100 /*numPileupBins*/, 100 /*maxPileupBin*/)
• MyDataPileupHistogram.root is the name of the output file, in which the profile will be stored.

Warning: this approach assumes that you are using 100% of the jets that have passed the trigger, which is not what happens in jet analyses.

Procedure

Get the pile-up profile in MC with the in-time pile-up (PileUp::intpu), and reweight the MC events so as to match the pile-up profile in data.

Systematic uncertainties

Vary the minimum-bias cross section with 4.6% (value from TWiki for Run 2).

Reweighting of simulation

<tt>getPUprofile</tt>

Extract the PU profile from MC and data.

<tt>applyPUprofCorrection</tt> and <tt>applyDiffPUprofCorrection</tt>

Apply correction on n-tuple, exactaly as shown by previous command (same header file is used).