Search for the Standard Model Higgs boson in the MET and b-jet Signature



Links

Home
Analysis Outline
Control Region Plots
Signal Region Plots
Results
 Contact Info

Analsis Outline

  We are analyzing Z-Higgs and W-Higgs associated production when the Z decays into two neutrinos, or the W decays leptonically but the electron or muon escape the detection. The data were collected with the CDF II detector at the Tevatron collider at Fermilab
  The final state we are interested in consists of two b-quark jets (from Higgs decay) and large Missing Transverse Energy (from Z/W decay). Many standard model process can produce the same final state, such as top pair production, single top, W/Z + jets, diboson production. These backgrounds are modelled using PYTHIA Monte Carlo simulation.
  QCD b-bbar production can mimic this signature due to severely mismeasured jets which appear to have large MET. Since the QCD heavy flavour production cross-section is orders of magnitude higher than that of the signal, it constitutes the biggest background in this search. Additionally light flavour jets can be falsely identified as b-jets (commonly referred to as "mistags"). We developed a technique which allows us to estimate both of these backgrounds in a unified manner using directly from data collected by CDF.
   We use two different algorithms to identify jets originating from b-quarks: "Secondary Vertex" and "Jet Probability" (SecVTX and JetProb). The events are split into three mutually exclusive tagging categories:
  - both jets are tagged by SecVTX,
  - one jet is tagged by SecVTX and the other by JetProb,
  - only one jet is tagged by SecVTX algorithm.
  We further increase the acceptance to signal by accepting events which contain a 3rd jet with ET>15 GeV. The third jet might be coming from hard radiation from the final state quarks, as well as from hadronic tau decays from WH->taunubb.
  We define several control regions to check our modeling of the data, check the performance of data-driven Multijet background modeling and validate the Monte-Carlo based background simulations.
  After achieving a good agreement in the control regions we need to define the signal region where we will perform the measurement. Since the QCD instrumental background has a distinct and different signature from that of signal and other backgrounds, we use an Artificial Neural Network to remove it. The Signal Region is defined by placing a cut on the output of this Neural Net. This approach has the advantage over traditional rectangular cuts in that it allows us to keep ~95% of signal while rejecting ~50% of backgrounds.
  To summarize the key improvements in this version of the analysis are:
  • Three mutually exclusive search channels, split by tagging categories. Single Tagged category adds ~10% to sensitivity.
  • Accept three jet events, where the 3rd jet is either a jet radiated off from a quark or a charged lepton. Adds sensitivity to WH->taunubb channel
  • Multijet background shape and normalization are estimated from data. Multijet normaliztion uncertainty reduced to <20%.
  • Jet energies are corrected using tracking information. Improves Dijet Mass resolution.
  • Neural Network based Signal Region definition to reject most of instrumental background. Improvement in signal acceptance with respect to cut-based selection
  Once the signal region is defined we train a Discriminant NN to further separate Multijet and ttbar backgrounds, and is used to search for the signal.
  Analyzing 2.1 fb-1 of CDF data we find a good agreement between expected background predictions and observed data, so we set an expected limit of 6.3 times the Standard Model prediction at the 95% CL for a Higgs boson mass of 115 GeV/c2, with an observed limit of 7.9*SM.


Created by Artur Apresyan.
For problems or questions regarding this website contact apresyan@fnal.gov
Last updated: 08/06/08.