Summary of the displaced SUSY parametrisation

The first period of data taking at the CERN LHC has ended and there is still no evidence of new physics. However, a large fraction of the searches assume that the new particle has a negligible lifetime and hence would decay very close to the interaction point. As these scenarios have been largely constrained, scenarios involving new particle with longer lifetimes have become more and more attractive. A search has been conducted that selects evens with an electron and muon which are displaced 1. No significant excess over the background expectation has been observed and limits on the 'Displaced Supersymmetry' scenario have been set. Even though this search sets limits on a specific model, it has been designed to be sensitive to a variety of BSM models.

On this page, information to facilitate the application of this search to other models of new physics is provided via a parametrisation of the signal acceptance based on Monte Carlo (MC) generator level quantities. The goal is to provide efficiency curves, which can be used on a set of simulated events for an arbitrary signal model involving exactly one electron and one muon that both have transverse impact parameter values between 0.02 cm and 2 cm in the final state, allowing the estimation of the expected signal yields.

The parametrisation has been validated (see below) by studying the reconstruction efficiency for both electrons and muons from the decay of a stop with mass between 200 and 1000 GeV and an average lifetime between 1 and 1000 mm. We obtain agreement within +/-25%, with a possible worsening for higher masses. Therefore, we strongly recommend that the use of this parametrisation be restricted to these mass and lifetime ranges, and that a 25% systematic uncertainty is assumed.

How to use the parametrisation

The user should follow all of the steps listed below.

  1. produce some generated events for the model of their choice
  2. apply all of the cuts listed below.
  3. reweight the surviving events with the four reconstruction and selection histograms and the trigger efficiency that are provided below. The five efficiencies are designed to be factorised.
  4. apply the d_0 cut on the electron and muon to match one of the 3 signal regions listed in the paper and compare the event yields with those observed in data (see, for example, table 1 of the paper).

List of cuts

Event passes standard FilterOutScraping cuts
One good primary vertex
generator electron coming from stop
generator electron with η < 2.5
generator electron with v0 < 4 cm
generator electron with vz < 30 cm
generator electron with pT > 25 CMS.GeV
generator muon coming from stop
generator muon with η < 2.5
generator muon with v0 < 4 cm
generator muon with vz < 30 cm
generator muon with pT > 25 CMS.GeV
event with exactly one generator electron
event with exactly one generator muon
generator electron and generator muon with opposite charge
generator electron-generator muon ∆R > 0.5
generator lectron-generator jet ∆R > 0.5
generator muon-generator jet ∆R > 0.5
generator electron with d0 < 2 cm
generator muon with d0 < 2 cm
Where vz (v0) is the position of the secondary vertex along the z-axis (in the the transverse plane).

Efficiency histograms and trigger efficiency

The following four efficiency histograms are provided. Tables corresponding to each figures can be found in HepData.

Figure Caption
el_reco.png Reconstruction efficiency of electron as function of d_0.
mu_reco.png Reconstruction efficiency of muon as function of d_0.
el_sel_rebinned.png Selection efficiency of electron as function of p_T.
mu_sel_rebinned.png Selection efficiency of muon as function of p_T.

The trigger efficiency is given as a single event weight. Its value is 95%.

Validation of the parametrisation

To evaluate the performance of the parametrisation, a closure test has been performed. For various signal samples, the ratio between the yield obtained with the parametrisation and the standard reconstruction is shown in a 2D plot.

Figure Caption
ClosureTest.png Ratio between the yield obtained with generator-level quantities and with the standard preselection for various signal samples.

Responsible: QuentinPython

Topic attachments
I Attachment History Action Size Date Who Comment
PDFpdf ClosureTest.pdf r1 manage 13.2 K 2015-02-27 - 21:56 QuentinPython  
PNGpng ClosureTest.png r1 manage 12.7 K 2015-02-27 - 21:56 QuentinPython  
PDFpdf el_reco.pdf r1 manage 15.4 K 2015-01-20 - 15:34 QuentinPython  
PNGpng el_reco.png r2 r1 manage 10.6 K 2015-02-27 - 21:18 QuentinPython  
PDFpdf el_sel_rebinned.pdf r1 manage 14.3 K 2015-01-20 - 15:34 QuentinPython  
PNGpng el_sel_rebinned.png r2 r1 manage 9.1 K 2015-02-27 - 21:19 QuentinPython  
PDFpdf mu_reco.pdf r1 manage 15.2 K 2015-01-20 - 15:34 QuentinPython  
PNGpng mu_reco.png r2 r1 manage 10.0 K 2015-02-27 - 21:20 QuentinPython  
PDFpdf mu_sel_rebinned.pdf r1 manage 14.4 K 2015-01-20 - 15:34 QuentinPython  
PNGpng mu_sel_rebinned.png r2 r1 manage 8.9 K 2015-02-27 - 21:20 QuentinPython  
Edit | Attach | Watch | Print version | History: r16 < r15 < r14 < r13 < r12 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r16 - 2016-05-11 - QuentinPython
    • 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 & 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback