Statistical models for CMS-SUS-24-007

Datacards repository for SUS-24-007 We are working on monoHiggs(bb) analysis in UL Run-II data (we analyze 2017 and 2018; and combine with published 2016 analysis at datacard level). This is a search for final state signature of large MET recoiling against a H(bb) candidate. The Higgs reconstruction is done in two categories, namely boosted and resolved where one AK8 jet and two AK4 jets are used. The background estimation is performed using per-METbin rateParams connecting the major backgrounds in signal and control regions. The signal extraction is performed by binned likelihood simultaneous fit. Detailed description can be found in analysis note AN-2024-044. We use Combine-v9 and cmsssw-el7 environment, and CMSSW_11_3_4 release. The input directory has datacard for one representative signal mass point for Baryonic-Z' model (1000GeV Zprime and 100GeV Chi) and combination of the two categories, and two years, namely dRun2_combined_rP_MCStat.txt. A sample datacard d17_boosted_TopMuCR_bin1.txt for 2017 Boosted category Top(mu) CR and METbin 1 is also added. For brief description of the signal extraction strategy using rateParam, please refer to slide 20 and 21 of the full status report https://indico.cern.ch/event/1446438/contributions/6096047/attachments/2915195/5115796/monoHbb_FullStatus_Aug23.pdf Eg. The rateParam ratett_Bcat_2018_METbin1 is to connect yield of ttbar process in Top(e) and Top(mu) CRs to SR in MET region 1 (250-350 GeV) for Boosted category in 2018. Checks Asimov Impacts To calculate impacts, following set of commands are use ulimit -s unlimited for r=0 combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -t -1 --expectSignal 0 --rMin -10 --doInitialFit --robustFit 1 -m 1000 -n _Run2Combined_sig0_Asimov_rP_MCStat combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -t -1 --expectSignal 0 --rMin -10 --doFits --robustFit 1 -m 1000 -n _Run2Combined_sig0_Asimov_rP_MCStat --parallel 32 combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -m 1000 -n _Run2Combined_sig0_Asimov_rP_MCStat -o impacts_Run2Combined_sig0_Asimov_rP_MCStat.json plotImpacts.py -i impacts_Run2Combined_sig0_Asimov_rP_MCStat.json -o impacts_Run2Combined_sig0_Asimov_rP_MCStat for r=1 combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -t -1 --expectSignal 1 --rMin -10 --doInitialFit --robustFit 1 -m 1000 -n _Run2Combined_sig1_Asimov_rP_MCStat combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -t -1 --expectSignal 1 --rMin -10 --doFits --robustFit 1 -m 1000 -n _Run2Combined_sig1_Asimov_rP_MCStat --parallel 32 combineTool.py -M Impacts -d dRun2_combined_rP_MCStat.root -m 1000 -n _Run2Combined_sig1_Asimov_rP_MCStat -o impacts_Run2Combined_sig1_Asimov_rP_MCStat.json plotImpacts.py -i impacts_Run2Combined_sig1_Asimov_rP_MCStat.json -o impacts_Run2Combined_sig1_Asimov_rP_MCStat FitDiagnostics and Pulls (Asimov) combine -M FitDiagnostics -d dRun2_combined_rP_MCStat.root --saveShapes --saveWithUncertainties -t -1 -n _Run2Combined_sig_Asimov_rP_MCStat --cminDefaultMinimizerStrategy 0 python diffNuisances.py --all fitDiagnostics_Run2Combined_sig_Asimov_rP_MCStat.root -g plots_Run2Combined_sig_Asimov_rP_MCStat.root >> ./fitResults_Run2Combined_sig_Asimov_rP_MCStat SR-masking (CRs-only fit) text2workspace.py dRun2_combined_rP_MCStat.txt --channel-masks -o dRun2_combined_rP_MCStat_SRmask.root cd maskSR/ combine -M FitDiagnostics -d dRun2_combined_rP_MCStat_SRmask.root --saveShapes --saveWithUncertainties --setParameters mask_d2017_Bcat_SR_METbin1=1,mask_d2017_Bcat_SR_METbin2=1,mask_d2017_Bcat_SR_METbin3=1,mask_d2018_Bcat_SR_METbin1=1,mask_d2018_Bcat_SR_METbin2=1,mask_d2018_Bcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin1=1,mask_d2017_Rcat_SR_METbin2=1,mask_d2017_Rcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin4=1,mask_d2017_Rcat_SR_METbin5=1,mask_d2018_Rcat_SR_METbin1=1,mask_d2018_Rcat_SR_METbin2=1,mask_d2018_Rcat_SR_METbin3=1,mask_d2018_Rcat_SR_METbin4=1,mask_d2018_Rcat_SR_METbin5=1 -n _Run2Combined_SRmask --cminDefaultMinimizerStrategy 0 python diffNuisances_Bonly.py --all fitDiagnostics_Run2Combined_SRmask.root -g plots_Run2Combined_SRmask_Bonly.root >> ./fitResults_Run2Combined_SRmask_Bonly Limits (Asymptotic expected limits) For each Chi mass, the following commands are run for each Zp mass. A representative command is shown below, combineTool.py -M T2W -i dRun2_combined_MZp1000MChi100GeV_rP_MCStat.txt -m 1000 -o workspaceRun2_combined_MZp1000MChi100GeV_rP_MCStat.root --parallel 20 combineTool.py -M AsymptoticLimits -d workspaceRun2_combined_MZp1000MChi100GeV_rP_MCStat.root -n .limit_MChi100GeV_Run2Combined -m 1000 --run blind --parallel 20 Then we collect the values in a json file combineTool.py -M CollectLimits higgsCombine.limit_MChi${ChiMass}GeV_Run2Combined.AsymptoticLimits.mH*.root -o limits_MChi${ChiMass}GeV_Combined_20172018.json Goodness of fit ulimit -s unlimited combine -M GoodnessOfFit -d dRun2_combined_rP_MCStat_SRmask.root --algo saturated -m 120.0 -n _result_Bonly_CRonly --setParameters mask_d2017_Bcat_SR_METbin1=1,mask_d2017_Bcat_SR_METbin2=1,mask_d2017_Bcat_SR_METbin3=1,mask_d2018_Bcat_SR_METbin1=1,mask_d2018_Bcat_SR_METbin2=1,mask_d2018_Bcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin1=1,mask_d2017_Rcat_SR_METbin2=1,mask_d2017_Rcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin4=1,mask_d2017_Rcat_SR_METbin5=1,mask_d2018_Rcat_SR_METbin1=1,mask_d2018_Rcat_SR_METbin2=1,mask_d2018_Rcat_SR_METbin3=1,mask_d2018_Rcat_SR_METbin4=1,mask_d2018_Rcat_SR_METbin5=1 combineTool.py -M GoodnessOfFit -d dRun2_combined_rP_MCStat_SRmask.root --algo saturated -n _result_Bonly_CRonly_Toys -t 500 --toysFrequentist -s 45239816 --setParameters mask_d2017_Bcat_SR_METbin1=1,mask_d2017_Bcat_SR_METbin2=1,mask_d2017_Bcat_SR_METbin3=1,mask_d2018_Bcat_SR_METbin1=1,mask_d2018_Bcat_SR_METbin2=1,mask_d2018_Bcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin1=1,mask_d2017_Rcat_SR_METbin2=1,mask_d2017_Rcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin4=1,mask_d2017_Rcat_SR_METbin5=1,mask_d2018_Rcat_SR_METbin1=1,mask_d2018_Rcat_SR_METbin2=1,mask_d2018_Rcat_SR_METbin3=1,mask_d2018_Rcat_SR_METbin4=1,mask_d2018_Rcat_SR_METbin5=1 --parallel 40 combineTool.py -M GoodnessOfFit -d dRun2_combined_rP_MCStat_SRmask.root --algo saturated -n _result_Bonly_CRonly_Toys -t 500 --toysFrequentist -s 76451209 --setParameters mask_d2017_Bcat_SR_METbin1=1,mask_d2017_Bcat_SR_METbin2=1,mask_d2017_Bcat_SR_METbin3=1,mask_d2018_Bcat_SR_METbin1=1,mask_d2018_Bcat_SR_METbin2=1,mask_d2018_Bcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin1=1,mask_d2017_Rcat_SR_METbin2=1,mask_d2017_Rcat_SR_METbin3=1,mask_d2017_Rcat_SR_METbin4=1,mask_d2017_Rcat_SR_METbin5=1,mask_d2018_Rcat_SR_METbin1=1,mask_d2018_Rcat_SR_METbin2=1,mask_d2018_Rcat_SR_METbin3=1,mask_d2018_Rcat_SR_METbin4=1,mask_d2018_Rcat_SR_METbin5=1 --parallel 40 hadd higgsCombine_result_Bonly_CRonly_Toys.GoodnessOfFit.mH120.Merged.root higgsCombine_result_Bonly_CRonly_Toys.GoodnessOfFit.mH120.76451209.root higgsCombine_result_Bonly_CRonly_Toys.GoodnessOfFit.mH120.45239816.root combineTool.py -M CollectGoodnessOfFit --input higgsCombine_result_Bonly_CRonly.GoodnessOfFit.mH120.root higgsCombine_result_Bonly_CRonly_Toys.GoodnessOfFit.mH120.45239816.root higgsCombine_result_Bonly_CRonly_Toys.GoodnessOfFit.mH120.76451209.root -m 120.0 -o gof_result_Bonly_CRonly_Run2Combined_toycomb.json plotGof.py gof_result_Bonly_CRonly_Run2Combined_toycomb.json --statistic saturated --mass 120.0 -o gof_result_Bonly_CRonly_Run2Combined_1000toys --title-right="B-only hypothesis(Combined Run2)"