Search for supersymmetry in final states with missing transverse energy and 0, 1, 2, 3, or ≥ 4 b jets in 8 TeV pp collisions

Public documentation

  • The result was published in Eur. Phys. J. C 73 (2013) 2568 and the preprint can be found at arXiv:1303.2985.
  • The Physics Analysis Summary on CDS (also found on INSPIRE), which describes the preliminary result for the HCP 2012 conference.
  • This public TWiki webpage summarises the analysis result and provides additional supporting material.

Abstract

A search for supersymmetry in final states with jets and missing transverse energy is performed in pp collisions at a centre-of-mass energy of 8 TeV. The data sample corresponds to an integrated luminosity of 11.7/fb collected by the CMS experiment at the LHC. In this search, a dimensionless kinematic variable, AlphaT, is used as the main discriminator between events with genuine and misreconstructed missing transverse energy. The search is performed in a signal region that is binned in the number of jets, the scalar sum of the transverse energy of these jets, and the number of jets identified as originating from a bottom quark. No excess of events over the standard model expectation is found. Exclusion limits are set in the parameter space of simplified models, with a special emphasis on compressed spectra and third-generation scenarios.

AlphaT distributions

The figures below show the AlphaT distributions from data for events that satisfy the baseline event selection criteria of the signal region, with the following exceptions. No requirement is made on AlphaT (which is nominally AlphaT > 0.55) or the number of reconstructed b-jets per event. The requirement HT > 375 GeV is imposed, and events are categorised according to the number of reconstructed jets in the event, as indicated in the figures. An inclusive and unbiased set of trigger conditions are used in order to show the full AlphaT distribution.

The analysis relies on data control samples and transfer factors (ratios) from simulation to estimate the contributions from both multijet and non-multijet backgrounds. However, for illustration, the expected yields from simulation for multijet events (green dash-dotted line), non-multijet backgrounds with genuine MET (dominated by W+jets, TTbar, and Znunu+jets; blue long-dashed line), the sum of these SM backgrounds (cyan solid line), and an example signal model (red dotted line) are also shown in the figures for comparison with the data (black solid circles with error bars). The statistical uncertainties for the multijet and SM expectations are represented by the hatched areas (visible only for statistically limited bins). The final bin contains all events with AlphaT > 3.

The figures highlight the ability of the AlphaT variable to discriminate between multijet events and all other SM or new physics processes with genuine MET in the final state. The expected yield for multijet events that satisfy AlphaT > 0.55, as given by simulation, is at the event-level and negligible with respect to all other SM backgrounds. (The final analysis result reaches the same conclusion, which relies on a data-driven technique to estimate the contamination from multijet events.) The distributions shown are inclusive with respect to HT and the number of reconstructed b-quark jets. Hence, the sensitivity to the example signal models shown can be significantly improved by categorising events according to HT and the number of reconstructed b-quark jets.

Link to figure Abbreviated Caption
AlphaT_le3j.png The AlphaT distribution for events satisfying the requirements defined above and containing either two or three reconstructed jets. The reference signal model (labelled D2) is the direct pair-production of bottom squarks, which results in a final state containing two b-quarks and two LSPs.

AlphaT_ge4j.png The AlphaT distribution for events satisfying the requirements defined above and containing at least four reconstructed jets. The reference signal model (labelled G2) is the gluino-mediated pair-production of (off-shell) bottom squarks, which results in a final state containing four b-quarks and two LSPs.

Closure tests and experimental systematic uncertainties

The method used to estimate the (non-multijet) standard model (SM) background contributions in the signal region relies on measurements made in three kinematically-similar data control samples, and the use of "translation factors" obtained from simulation. The signal region and data control samples are binned according to the scalar sum of the jet transverse energies, HT, the number of reconstructed jets per event, Njet, and the number of reconstructed b-quark jets per event, Nb. Translation factors are constructed separately for each data control sample and per (HT,Njet,Nb) bin.

Systematic uncertainties are assigned to the translation factors obtained from simulation, which account for theoretical uncertainties and limitations in the simulation modelling of kinematics and instrumental effects. The magnitudes of these systematic uncertainties are determined from data with a representative set of closure tests, in which yields from one of the three control samples, along with the corresponding translation factors obtained from simulation, are used to predict the yields in another control sample. The samples are selected to ensure that potential signal contamination in a wide variety of SUSY models, including those considered in this analysis, are negligible. Therefore, the closure tests carried out both between and within control samples probe the properties of the relevant SM backgrounds. In particular, the closure tests address the modelling in simulation of the relative contributions of the main SM backgrounds (W + jets, top and Z + jets), the kinematics of important observables such as AlphaT, and the reconstruction of b-quark jets.

Closure tests for the two jet multiplicity bins used in the analysis (2 ≤ Njet ≤ 3, Njet ≥ 4) are shown in the plots below. In each plot, there are eight sets of closure tests, where each set comprises (up to) eight individual closure tests, one per HT bin. All error bars represent statistical uncertainties only. The first three sets are carried out within the muon + jets sample. The first set probes the modelling of the AlphaT distribution in events with significant, genuine MET (circles), whilst the second (squares) and third (triangles) sets examine the relative composition between W + jets and top events, as well as the modelling of the reconstruction of b-quark jets. These are particularly important for the estimation of the W + jets and top component of the SM background. The fourth (crosses) and fifth (stars) sets are performed between different control samples and are important for establishing a consistent estimation of the irreducible component of the Z boson decaying to two neutrinos. The fourth set connects the muon + jets and di-muon + jets control samples testing the relative contributions of Z + jets to W + jets (and top) events. The fifth set establishes the consistency between the Z boson (decaying to two muons) + jets and photon + jets samples. A further three sets of closure tests (inverted triangles, diamonds, asterisks) probe the simulation modelling of the jet multiplicity distribution in data for each of the three control samples in turn. In all cases, the closure tests are overlaid on top of grey bands that represent the systematic uncertainties that are assigned to the translation factors, which are as large as 30% at high HT.

Link to figure Abbreviated Caption
syst-le3j.png Closure tests between data control samples for events containing two or three jets. Systematic uncertainties on the translation factors are indicated by the shaded bands.

syst-ge4.png Closure tests between data control samples for events containing at least four jets. Systematic uncertainties on the translation factors are indicated by the shaded bands.

The result: observed yields and standard model expectations

A likelihood model of the observations in the signal region and three data control samples is used to obtain a consistent prediction of the standard model (SM) background, and to test for the presence of a variety of signal models. The likelihood function contains terms that describe the yields in each of the data samples that are binned according to the scalar sum of the jet transverse energies, HT, the number of reconstructed jets per event, Njet, and the number of reconstructed b-quark jets per event, Nb. There are eight bins in HT (with lower bin boundaries in the range from 275 GeV to 875 GeV), two Njet bins (2 ≤ Njet ≤ 3, Njet ≥ 4), and five Nb bins (exactly zero, one, two, three, or at least four b-quark jets per event).

In each (HT,Njet,Nb) bin of the signal region, the observation is modelled as Poisson-distributed about the sum of a SM expectation and a potential signal contribution. The components of this SM expectation are related to the expected yields in the corresponding (HT,Njet,Nb) bins of the control samples via translation factors derived from simulation. Signal contributions in each of the data samples are considered, though the only significant contribution occurs in the signal region and not the control samples. The systematic uncertainties associated with the translations are accounted for with nuisance parameters, the measurements of which are treated as log-normally distributed. Since when requiring at least two b-quark jets, the dominant SM background arises from top events, only the muon + jets control sample is used in the likelihood to determine the total contribution from all (non-multijet) SM backgrounds in the signal region. The contribution from the multijet background is modelled to have an exponential dependence on HT, but is expected to be negligible.

In order to test the compatibility of the observed yields with the expectations from SM processes only, the likelihood function is maximised over all fit parameters. Eight categories of (Njet,Nb) are considered, as summarised in the table below.

Njet 2-3 2-3 2-3 ≥ 4 ≥ 4 ≥ 4 ≥ 4 ≥ 4
Nb 0 1 2 0 1 2 3 ≥ 4

No significant excess above the SM expectation is observed in the hadronic signal region, and the control samples are well described by the SM hypothesis.

Comparisons of the observed yields and the SM expectations in bins of HT for these eight categories of (Njet,Nb) are shown in the plots below. Each plot shows the observed event yields in data (black dots) and the expectations and their uncertainties, as determined by the simultaneous fit, for all SM processes (light blue solid line with dark blue bands). For illustrative purposes only, the expected yields from an example signal model is also shown (magenta solid line), superimposed on top of the SM expectation.

Tabular form of the fit result

A comparison of the observed yields and the SM expectations in bins of HT for these eight categories of (Njet,Nb) are also available in tabular form here.

Signal region

Link to figure Abbreviated Caption
0b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring 2 ≤ Njet ≤ 3 and Nb = 0. The example signal model comprises pair-produced 600 GeV squarks, each decaying to a quark and a 250 GeV neutralino LSP. The model assumes eight first- and second-generation squarks with degenerate masses.
1b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring 2 ≤ Njet ≤ 3 and Nb = 1. The example signal model comprises pair-produced 500 GeV bottom squarks, each decaying to a bottom quark and a 150 GeV neutralino LSP.
2b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring 2 ≤ Njet ≤ 3 and Nb = 2. The example signal model comprises pair-produced 500 GeV bottom squarks, each decaying to a bottom quark and a 150 GeV neutralino LSP.
0b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring Njet ≥ 4 and Nb = 0. The example signal model comprises pair-produced 700 GeV gluinos, each decaying to a quark-antiquark pair and a 300 GeV neutralino LSP.
1b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring Njet ≥ 4 and Nb = 1. The example signal model comprises pair-produced 400 GeV top squarks, each decaying to a top quark and a 0 GeV neutralino LSP.
2b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring Njet ≥ 4 and Nb = 2. The example signal model comprises pair-produced 400 GeV top squarks, each decaying to a top quark and a 0 GeV neutralino LSP.
3b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring Njet ≥ 4 and Nb = 3. The example signal model comprises pair-produced 900 GeV gluinos, each decaying to a bottom quark-antiquark pair and a 500 GeV neutralino LSP.
ge4b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the signal region when requiring Njet ≥ 4 and Nb ≥ 4. The example signal model comprises pair-produced 850 GeV gluinos, each decaying to a top quark-antiquark pair and a 250 GeV neutralino LSP.

Muon + jets control sample

Link to figure Abbreviated Caption
0b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 0.
1b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 1.
2b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 2.
0b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring Njet ≥ 4 and Nb = 0.
1b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring Njet ≥ 4 and Nb = 1.
2b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring Njet ≥ 4 and Nb = 2.
3b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring Njet ≥ 4 and Nb = 3.
ge4b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the muon + jets control sample when requiring Njet ≥ 4 and Nb ≥ 4.

Di-muon + jets control sample

Link to figure Abbreviated Caption
0b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the di-muon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 0.
1b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the di-muon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 1.
0b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the di-muon + jets control sample when requiring Njet ≥ 4 and Nb = 0.
1b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the di-muon + jets control sample when requiring Njet ≥ 4 and Nb = 1.

Photon + jets control sample

Link to figure Abbreviated Caption
0b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the photon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 0.
1b_le3j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the photon + jets control sample when requiring 2 ≤ Njet ≤ 3 and Nb = 1.
0b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the photon + jets control sample when requiring Njet ≥ 4 and Nb = 0.
1b_ge4j Comparison of the observed yields and SM expectations given by the simultaneous fit in bins of HT for the photon + jets control sample when requiring Njet ≥ 4 and Nb = 1.

Interpretations of the result with simplified models

Signal efficiency times acceptance

The figures below indicate the signal efficiency times acceptance as a function of the mass of the produced sparticle (gluino or squark) and the LSP mass. The efficiency times acceptance value for each model point is calculated by summing over all HT bins and the (Njet,Nb) event categories considered for a given model (specified in the table below). The figure content is also available in digital form (the ROOT file format) via the link above each figure.

Model Njet Nb Topology Figure with embedded link to PDF file Abbreviated Caption
T1 ≥ 4 0 Figure content in digital form (ROOT file)
T1
Signal efficiency times acceptance versus model spectrum for pair-produced gluinos, each decaying to a quark-antiquark pair and a neutralino LSP.
T2 2-3 0 Figure content in digital form (ROOT file)
T2
Signal efficiency times acceptance versus model spectrum for pair-produced squarks, each decaying to a quark and a neutralino LSP.
T2bb 2-3 1, 2 Figure content in digital form (ROOT file)
T2bb
Signal efficiency times acceptance versus model spectrum for pair-produced bottom squarks, each decaying to a bottom quark and a neutralino LSP.
T2tt ≥ 4 1, 2 Figure content in digital form (ROOT file)
T2tt
Signal efficiency times acceptance versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP.
T1tttt ≥ 4 2, 3, ≥ 4 Figure content in digital form (ROOT file)
T1tttt
Signal efficiency times acceptance versus model spectrum for pair-produced gluinos, each decaying to a top quark-antiquark pair and a neutralino LSP.
T1bbbb ≥ 4 2, 3, ≥ 4 Figure content in digital form (ROOT file)
T1bbbb
Signal efficiency times acceptance versus model spectrum for pair-produced gluinos, each decaying to a bottom quark-antiquark pair and a neutralino LSP.

Observed upper limits for the production cross section and excluded mass regions

The figures below show the excluded cross section versus model spectrum for various simplified models. The colour scale indicates the observed cross section upper limit (95% CL) as a function of the mass of the produced sparticle (gluino or squark) and the LSP mass. The point-to-point fluctuations are due to the finite number of pseudo-experiments used to determine the observed upper limit. The thick solid black line indicates the observed exclusion region assuming nominal NLO+NLL SUSY production cross section for pair-produced gluinos in the limit of very massive squarks (or vice versa). The thin black lines represent the observed excluded region when varying the production cross section by its theoretical uncertainty. The dashed purple lines indicate the median (thick line) +/- 1 sigma (thin lines) expected exclusion regions. Only certain categories of (Njet,Nb) are considered for each simplified model, as summarised below. The figure content is also available in digital form (the ROOT file format) via the link above each figure.

Model Njet Nb Topology Link to figure Abbreviated Caption
T1 ≥ 4 0 Figure content in digital form (ROOT file)
T1
Exclusion cross-section versus model spectrum for pair-produced gluinos, each decaying to a quark-antiquark pair and a neutralino LSP.
T2 2-3 0 Figure content in digital form (ROOT file)
T2
Exclusion cross-section versus model spectrum for pair-produced squarks, each decaying to a quark and a neutralino LSP. The two sets of exclusion contours correspond to the production of eight first- and second-generation squarks with degenerate masses or only a single light squark.
T2bb 2-3 1, 2 Figure content in digital form (ROOT file)
T2bb
Exclusion cross-section versus model spectrum for pair-produced bottom squarks, each decaying to a bottom quark and a neutralino LSP.
T2tt ≥ 4 1, 2 Figure content in digital form (ROOT file)
T2tt
Exclusion cross-section versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP. Only the expected excluded regions are indicated, as no observed exclusion is possible for the considered mass parameter space.
(The 1D limit plots for four different LSP masses can be found below.)
T1tttt ≥ 4 2, 3, ≥ 4 Figure content in digital form (ROOT file)
T1tttt
Exclusion cross-section versus model spectrum for pair-produced gluinos, each decaying to a top quark-antiquark pair and a neutralino LSP.
T1bbbb ≥ 4 2, 3, ≥ 4 Figure content in digital form (ROOT file)
T1bbbb
Exclusion cross-section versus model spectrum for pair-produced gluinos, each decaying to a bottom quark-antiquark pair and a neutralino LSP.

T2tt simplified model for fixed LSP masses

The figures below show the excluded cross sections versus top-squark mass for the T2tt simplified model, in which pair-produced top squarks decay to a top quark and the LSP of fixed mass (as indicated below). The observed upper limit (95% CL) on the production cross section is shown as a function of the top-squark mass (solid line), along with the expected upper limit and +/- 1 sigma experimental uncertainties (long-dashed line with shaded band), and the NLO+NLL top-squark pair-production cross section and theoretical uncertainties (dotted line with shaded band). All interpretations for this model are performed using the event categories that satisfy Njet ≥ 4 and Nb = 1 or Nb-jet = 2.

LSP mass Topology Link to figure Abbreviated Caption
0 GeV T2tt_mlsp0 Exclusion cross-section versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP of mass 0 GeV.
50 GeV T2tt_mlsp50 Exclusion cross-section versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP of mass 50 GeV.
100 GeV T2tt_mlsp100 Exclusion cross-section versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP of mass 100 GeV.
150 GeV T2tt_mlsp150 Exclusion cross-section versus model spectrum for pair-produced top squarks, each decaying to a top quark and a neutralino LSP of mass 150 GeV.

Additional material

LHE files for benchmark signals

We provide the LHE files for six benchmark signal points:

Topic attachments
I Attachment History Action Size Date Who Comment
Unknown file formatlhe T1_700_300.lhe r1 manage 13725.2 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Unknown file formatlhe T1bbbb_900_500.lhe r1 manage 13764.3 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Unknown file formatlhe T1tttt_850_250.lhe r1 manage 170951.8 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Unknown file formatlhe T2_600_250.lhe r1 manage 11186.2 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Unknown file formatlhe T2bb_500_150.lhe r1 manage 11205.7 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Unknown file formatlhe T2tt_400_0.lhe r1 manage 106694.0 K 2015-03-09 - 09:48 NadjaStrobbe benchmark LHE file
Edit | Attach | Watch | Print version | History: r27 < r26 < r25 < r24 < r23 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r27 - 2016-11-09 - RobBainbridge
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback