MSSM Higgs production in association with b-quarks - semi leptonic

his is a condensed description with plots for the analysis CMS-PAS-HIG-12-027

Abstract

This paper presents a search of neutral Supersymmetric Higgs particles produced in association with two spectator b-quarks and decaying to pairs of b-quarks. The data used correspond to 4.8 fb-1 collected by the CMS experiment at the LHC during the 2011 run, at a center of mass energy of 7 TeV. The data passed a semi-leptonic trigger that required a muon in the final state. Two different data driven methods to predict the background are described, yielding low statistical and systematic uncertainties. The results are presented in the framework of MSSM, in the m_h^max scenario.

Main results

Measurements of this search are the 95% CL limits on the MSSM cross section and the 95% exclusion limit in the MSSM tan beta - mA plane

Tables from the PAS HIG-12-027

Table 1:

Data reduction after each selection cut. The total integrated luminosity used is Ldt [ pb−1 ] = 4805.7

Cut	Events
All	16732273
μ pT > 15 GeV/c	9739139
#jets >= 3	4511327
∆Rij >= 1	3505584
CSV (1st − jet ) > 0.8	1932135
CSV (2nd − jet) > 0.8	813685
μ in 1st or 2nd jet	785940
CSV (3rd − jet) > 0.7	60195

Table 2:

Events passing all analysis cuts for MC simulated background samples. The quoted errors are statistical only.

Process	events
Multijet QCD	72043±747
tt + jets	303±5.8
Z → bb	540±70
ZZ	1.84±0.07
WZ	0.44±0.05
WW	0.03±0.03

From PAS text

The relevant background processes for this analysis are studied using simulated events, produced for tt by the MADGRAPH [14] Monte Carlo (MC) generator and with PYTHIA [15] for the other processes. The Z → bb yield is rescaled to match the MADGRAPH jets multiplicity. All simulated datasets are then processed using GEANT4 [16] with a detailed CMS detector simulation, including the PileUp events as observed during 2011 data taking, and finally reconstructed with the same reconstruction program used for data. The expected events for each process are shown in Tab. 2: each process is normalized using the measured cross section to the total integrated luminosity used, with the exception of multijet QCD background, for which the nominal PYTHIA cross section has been used.

Figures from the PAS HIG-12-027

Figure	Link	Description
	DiscriminatorMC.png DiscriminatorMC.pdf	Figure 1: Distribution of the discriminating variable used for M H ≤ 180 GeV/c2 as defined in the text for multijet QCD background events (black), tt (blue), Z → bb (green) and SUSY Higgs 2 (red). All distributions are normalized to unity. In the analysis, signal for m H = 120 GeV/c the multijet QCD background is the largest one by far, and tt → W + bW − b and Z → bb are expected to be small, as shown in Tab. 2.
	FitMass.pdf FitMass.png FitMJPro.pdf FitMJPro.png FitJPro.pdf FitJPro.png	Figure 2: Example of a fit for one bin of the ET , eta parametrization. The red, blue, and green histograms are respectively the templates for b, c, and light quarks. Dots represent data. Left [FitMass] and center [FitMJPro]: two-dimensional fit to Tag Mass and JetBProbability. Right [FitJPro]: fit to JetBProbability only when the Tag Mass is not defined.
	Minv01.pdf Minv01.png	Figure 3: Invariant mass of the two leading jets, for MC and Data and for control and signal re-gions. Prediction (blue histogram), data (red dots) are overlaid. The prediction is normalized to data, and the normalization scale factor is also shown for each plot (Seen/Pred). Also a hypothesis for a SUSY higgs signal at Mh = 120 GeV/c2 with a production cross section corresponding to tan beta = 30 (green histogram) is overlaid.
	Aver_shaperatio_Le200.pdf Aver_shaperatio_Le200.png Aver_shaperatio_Gt200.pdf Aver_shaperatio_Gt200.png	Figure 4: Background predictions from the B-matrix and the Hyperball method compared for low mass region M H ≤ 180 GeV/c2 [Aver_shaperatio_Le200] and high mass region M H > 180 GeV/c2 [Aver_shaperatio_Gt200].
	Minv01_SignalRegion_Le200.pdf Minv01_SignalRegion_Le200.png	Figure 5: Data (red) and predicted background (blue) in the signal region, for low mass range (M H ≤ 180 GeV/c2 ); the expected signal for different M H and for tan β = 30 is also plotted. Linear scale on the left, logarithmic on the right. The difference between data and predicted background is also shown.
	Minv01_SignalRegion_Gt200.pdf Minv01_SignalRegion_Gt200.png	Figure 6: Data (red) and predicted background (blue) in the signal region, for low mass range (M H > 180 GeV/c2 ); the expected signal for different M H and for tan β = 30 is also plotted. Linear scale on the left, logarithmic on the right. The difference between data and predicted background is also shown.
	CLs_Xsec_withSyst.pdf CLs_Xsec_withSyst.png CLs_Xsec_Logy_withSyst.pdf CLs_Xsec_Logy_withSyst.png	Figure 7: Observed and expected upper limit for the cross section times branching ratio for 95% confidence level, with linear [CLs_Xsec_withSyst] and logaritmic [CLs_Xsec_Logy_withSyst] scale, with the inclusion of statistical and systematic uncertainties. The low and high mass regions are shown separately.
	CLs_Tanb_withSyst.pdf CLs_Tanb_withSyst.png	Figure 8: Observed and expected upper limit with 90% confidence level in the MSSM plane (tan β, M A ), including the statistical and systematical uncertainties, in the mhmax benchmark scenario as defined in the text. The low and high mass regions are shown separately.
	CLs_Tanb_withSyst_CdfD0.pdf CLs_Tanb_withSyst_CdfD0.png	Additional Figure: Figure 49: Observed and expected upper limit with 90% confidence level in the MSSM plane ( M A , tan β), for m_h^max benchmark scenario, with μ = −200 GeV/c2 , including the statistical and systematical uncertainties, overlaied to the previously published CDF and D0 results [7, 8]

Conclusion

A search of neutral supersymmetric Higgs particles decaying into pairs of b-quarks, and pro- duced in association with two further b-quarks, has been presented. The data were collected during 2011 by the CMS experiment at the LHC, corresponding to a total integrated luminosity of 4.8 fb−1 , with the use of a semileptonic trigger. Two data driven predictions of the large and dominating multijet QCD background have been developed.

The data shows no significant excess with respect to the expected SM background, and a limit for Higgs cross section times branching ratio as a function of M A is determined. These results are also interpreted in the framework of MSSM as a function of ( M A , tan β), in the m_h^max [18, 19] scenario, excluding a region of phase space previously unexplored for this final state.

-- StefanoLacaprara - 28-Jun-2012