Bootstrapping mixed data

Plan

  • use data*_v* files to map event, run -> (h1event, h1run, h2event, h2run )
    • /store/user/jda102/condor/ZH4b/ULTrig/data2018_v0/picoAOD_3bDvTMix4bDvT_v0_newSBDef.root
  • Make dictionary ala synthetic datasets of events to SvB
  • Need another dictionary of events to hemievent pairs
  • Assign hemievent pairs to SvB
  • Fluctuate weights.
  • Compute combined weights

Logs

3 April 2025 Thursday

2 April 2025 Wednesday

  • cmslpc304
  • sed -e "s|run_dilep_ttbar_crosscheck: true|run_dilep_ttbar_crosscheck: false|" analysis/metadata/HH4b.yml > output/analysis_test_mixed_job/HH4b.yml
  • python runner.py -o mixedData_wSvB.coffea -d mixeddata -p analysis/processors/processor_HH4b.py -y 2016 2017 2018 -op output/analysis_test_mixed_job -m metadata/datasets_HH4b.yml -c output/analysis_test_mixed_job/HH4b.yml --condor
  • Setting up to run on the mixed data
    • add datasets to the input yml file
    • modify the runner to run on datamixed
    • adding h1 and h2 event, run to output files
  • Testing
    • python runner.py -t -o testDataMixed_wSvB.coffea -d datamixed -p analysis/processors/processor_HH4b.py -y 2016 -op output/analysis_test_mixed_job -m metadata/datasets_HH4b_datamixed.yml -c output/analysis_test_mixed_job/HH4b.yml --condor
  • running on condor
    • python runner.py -o dataMixed_hNumbers.coffea -d datamixed -p analysis/processors/processor_HH4b.py -y 2016 2017 2018 -op output/analysis_test_mixed_job -m metadata/datasets_HH4b_datamixed.yml -c output/analysis_test_mixed_job/HH4b.yml --condor
  • Script to prepare dataset
    • =python jetclustering/mixingcorrelationprepdata.py output/analysistestmixedjob/mixedDatawSvB.coffea output/analysistestmixedjob/dataMixedhNumbers.coffea =
  • Made merged dataset
  • [>>] Bootstrap with hemisphere weights

Follow-ups

Links:

Bootstrapping Synthetic Data Stat Uncertainties

202504090843

Backlinks