Search for supersymmetry using hadronic top quark tagging in 13 TeV pp collisions (SUS-16-050)

Further information

This analysis is documented in arXiv:1710.11188 and the public webpage for this analysis is http://cms-results.web.cern.ch/cms-results/public-results/preliminary-results/SUS-16-050/index.html.

Technical plots for CMS speakers

Top Tagger

Show Details Hide Details

Figure	Caption
	The discriminant distribution of the MVA trained for the triple-jet category top tagger in semileptonic ttbar MC after the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet. The results are shown separately for top candidates matched to generator level top quarks and unmatched candidates.
	The misidentification rate of the top tagger used in SUS-16-050 in comparison with the previous cut-base version of the top tagger used in SUS-16-009. The misidentification rate is calculated on an event-wise basis using Z→νν MC as the ratio of events with at least one reconstructed top quark to the total number of events as a function of pTmiss. The results are produced with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.
	The trijet candidate mass variable which is used as input to the top tagger MVA. The distribution is shown for 4 different categories: the candidates passing/failing the MVA random forest discriminator are shown in red/blue respectively, while those matched to a generator-level top quark vs unmatched are shown in dotted green/black respectively. The results are produced from a ttbar MC sample generated with one leptonic and one hadronic decaying top with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.
	The invariant mass of the second and third highest momentum jets in the trijet rest frame which is used as input to the top tagger MVA. The second and third highest momentum jets sorted in the rest frame should form the W candidate. The distribution is shown for 4 different categories: the candidates passing/failing the MVA random forest discriminator are shown in red/blue respectively, while those matched to a generator level top quark vs unmatched are shown in dotted green/black respectively. The results are produced from a ttbar MC sample generated with one leptonic and one hadronic decaying top with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.
	The angular separation in the trijet rest frame between the second and third highest momentum jets in the trijet rest frame which is used as input to the top tagger MVA. The distribution is shown for 4 different categories: the candidates passing/failing the MVA random forest discriminator are shown in red/blue respectively, while those matched to a generator level top quark vs unmatched are shown in dotted green/black respectively. The results are produced from a ttbar MC sample generated with one leptonic and one hadronic decaying top with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.
	The momentum of the highest momentum jet in the trijet rest frame which is used as input to the top tagger MVA. The distribution is shown for 4 different categories: the candidates passing/failing the MVA random forest discriminator are shown in red/blue respectively, while those matched to a generator-level top quark vs unmatched are shown in dotted green/black respectively. The results are produced from a ttbar MC sample generated with one leptonic and one hadronic decaying top with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.
	The combined secondary vertex (CSV) b-tagging discriminator of the highest momentum jet in the trijet rest frame which is used as input to the top tagger MVA. The discontinuity around 0.84 is due to the requirement that no top can have more than one b-tagged jet. The distribution is shown for 4 different categories: the candidates passing/failing the MVA random forest discriminator are shown in red/blue respectively, while those matched to a generator level top quark vs unmatched are shown in dotted green/black respectively. The results are produced from a ttbar MC sample generated with one leptonic and one hadronic decaying top with the following requirements: pTmiss > 250 GeV, no electron or muon with pT > 10 GeV, at least 4 jets with pT > 30 GeV, the Δφ between the three leading pT jets and the pTmiss must be greater than 0.3, and at least one b-tagged jet.

Background prediction from Monte Carlo

Show Details Hide Details

Figure	Caption
	Background composition in 0-lepton search region in bins of the number of b-tagged jets and the number of top-tagged candidates from the simulated events. The selection of search region is described in section 8 of the paper. Each background are normalized to its SM cross section.

Background prediction from analysis

Show Details Hide Details

Figure	Caption
	The distribution of HT is shown for data and the final pre-fit background prediction in the signal region. The selection is described in section 8 of the paper. Four example signal points, normalized to their expected cross sections, are also shown for comparison. The choice of binning is chosen to approximately match the binning used in the primary search bins of the analysis.
	The distribution of pTmiss is shown for data and the final pre-fit background prediction in the signal region. The selection is described in section 8 of the paper. Four example signal points, normalized to their expected cross sections, are also shown for comparison. The choice of binning is chosen to approximately match the binning used in the primary search bins of the analysis.
	The distribution of MT2 is shown for data and the final pre-fit background prediction in the signal region. The selection is described in section 8 of the paper. Four example signal points, normalized to their expected cross sections, are also shown for comparison. The choice of binning is chosen to approximately match the binning used in the primary search bins of the analysis.
	The distribution of Nb is shown for data and the final pre-fit background prediction in the signal region. The selection is described in section 8 of the paper. Four example signal points, normalized to their expected cross sections, are also shown for comparison. The choice of binning is chosen to approximately match the binning used in the primary search bins of the analysis.
	The distribution of Nt is shown for data and the final pre-fit background prediction in the signal region. The selection is described in section 8 of the paper. Four example signal points, normalized to their expected cross section, are also shown for comparison. The choice of binning is chosen to approximately match the binning used in the primary search bins of the analysis.

Event Display

Show Details Hide Details

Figure	Caption
	Event display for a SUSY candidate event (from Run 274250, Event number 425322875) with two reconstructed top-tagged candidates (one is in di-jet category, the other is in tri-jet category) passed the search region selection, in view of the CMS detector. The momenta of the AK4 jets that tagged as b-jet in tri-jet category are marked in green, while other AK4 jets are marked in blue. The momentum of the AK8 jets are marked in yellow and missing transverse momentum is marked in purple.
	Event display for a SUSY candidate event (from Run 274969, Event number 766814305) with two reconstructed top-tagged candidates (one is in mono-jet category, the other is in tri-jet category) passed the search region selection, in view of the CMS detector. The momenta of the AK4 jets that tagged as b-jet in tri-jet category are marked in green, while other AK4 jets are marked in blue. The momentum of the AK8 jets are marked in yellow and missing transverse momentum is marked in purple.