Skip to content

Preliminary pileup reweighting, weights in TnP processor, bug fixes in TnP selection

Florian Alexander Mausolf requested to merge fmausolf/HiggsDNA:pileup_reweighting into master

1. Inclusion of pileup reweighting into the base and TagandProbe processors.

Since for run-3, there is no central pileup json available yet, I followed the description from https://twiki.cern.ch/twiki/bin/viewauth/CMS/PileupJSONFileforData to produce two pileup histograms, one for 2022F and one for the full 2022 data taking. These are now located at higgs_dna/metaconditions/pileup_histograms/.

They were produced starting from the corresponding certified Golden JSONs, https://cms-service-dqmdc.web.cern.ch/CAF/certification/Collisions22/Cert_Collisions2022_355100_362760_Golden.json (full 2022 data) and https://cms-service-dqmdc.web.cern.ch/CAF/certification/Collisions22/Cert_Collisions2022_eraF_360390_362167_Golden.json (data 2022F), with the following commands: a) brilcalc lumi --xing -b "STABLE BEAMS" -i Cert_Collisions_2022FULL_or_ERA_F_Golden.JSON --xingTr 0.1 -o lumi_DCSONLY.csv --normtag /cvmfs/cms-bril.cern.ch/cms-lumi-pog/Normtags/normtag_BRIL.json b) makePileupJSON.py lumi_DCSONLY.csv pileup_JSON.txt c) pileupCalc.py -i Cert_Collisions_2022FULL_or_ERA_F_Golden.JSON --inputLumiJSON pileup_JSON.txt --calcMode true --minBiasXsec 80000 --maxPileupBin 100 --numPileupBins 100 MyDataPileupHistogram.root

Note that here, I used 80mb as inelastic proton scattering cross-section. Keep in mind: if there will not be a centrally provided pileup JSON which we could use to replace this, we need to investigate if this cross-section is ideal and add systematic variations on our own. However, the adaptions needed in HiggsDNA should be minor.

2. Inclusion of weights into tnp processor

weights are now added to the tnp_candidates, keeping into account that due to the TnP combinations, one event can be counted two times as tag&probe pair. Weight corrections and systematics still need to be ported to this processor, should work smoothly by following what is present in base.

3. Bug fixes in TnP selection

Electron veto cut was applied previously, since the photon_preselection was ported to the external file. Added flag in the photon_preselection whether to apply veto cut or not.

Corrected probe rel. iso. cut, where tag pT was used.

Merge request reports