mass. Therefore, it is used to combine the result with the 7 TeV dataset giving a diboson signal strengths ofµV Z = 0.74±0.09(stat.)±0.14(syst.).
The observed significance of the V Z signal is 4.9σ, to be compared to an expected significance of 6.3σ. Additional measurements where the signal strengths are floated independently for the 7 and 8 TeV datasets and for the three lepton channels are carried out as crosschecks. Figure 8.2 shows the results.
(a) (b)
Figure 8.2: The fitted values of the diboson signal strengthµV Z for (a) the 7 TeV, 8 TeV and combined datasets, and (b) for the three lepton channels separately and combined, for the combined dataset. The MVA is used for the 8 TeV data. The individualµV Z values for the lepton channels are obtained from a simultaneous fit with the signal strength for each channel floating independently.
As a crosscheck as well, additional fits are performed with two signal strengths left freely floating for the V Z and Higgs processes in order to check any possible bias. The test uses the same final discriminants as for the Higgs boson analysis search on the 8 TeV dataset. The results of this test do not show any variation in the signal strength compare to the nominal procedure where the Higgs boson is normalized to its SM cross section value with an uncertainty of 50 %. Correlations betweenµV Z and µV H, obtained from the correlation matrices given by the fit, are 9 % in the dijet mass analysis and -3 % in the MVA. The reason for the low correlation between the signal strengths is the different shape of the pVT distributions for V Z and Higgs boson events. In addition and as previously mentioned, the MVA analysis discriminates better between processes.
8.2 Dijet mass analysis results
Final results using the dijet mass analysis strategy on the 8 TeV data are presented in this Section. A full set of distributions for the 0-leptonmbband
CHAPTER 8. RESULTS
the MV1c discriminants, with the background normalizations and nuisance parameters adjusted from a fit to data, were already presented in Section 7.1.
Figures 8.3 and 8.4 show a summary ofmbbdistributions for the 0-, 1- and 2-lepton channels, where theb-jet categories with the highest purity, 2-Medium and 2-Tight, as well as, the pVT intervals with higher values are combined.
In general, good agreement between data and expected number of events is presented.
The 95 % CL upper limits on the cross section times branching ratio for pp → (W/Z)(H → b¯b) in the Higgs boson mass range 110–140 GeV is presented in Figure 8.5. The Figure shows how the analysis is most sensitive to the presence of a Higgs signal in the low mass range. The reason is that the branching ratio of the Higgs boson decaying into b-quarks decreases with its mass. The Figure also presents differences in all the mass range between the expected limit in the absence of a signal and the corresponding observed values. This indicates the presence of an excess of events above the expected background. In particular, the observed limit in the absence of signal obtained formH = 125 GeV is 2.2 times the SM value, to be compared to an expected limit of 1.02.
The probability p0 associated with the background-only hypothesis at the Higgs boson mass of 125 GeV is 1.3 %. This corresponds to an excess observed at this mass with a significance of 2.2σ, to be compared to an expectation of 1.9σ. The p0 values for the Higgs mass range 110–140 GeV obtained with the 8 TeV dataset is presented in Figure 8.6.
The fitted signal strength value for the dijet mass 8 TeV analysis for a 125 GeV Higgs boson mass isµ= 1.23±0.44(stat.)±0.41(syst.). Additional measurements are also performed, where the signal strengths are floated in-dependently for theW H and ZH production processes or the three lepton channels. Figures 8.7 and 8.8 show the fitted µvalues for these additional measurements compared to the nominal dijet result. The compatibility be-tween the signal strengths in the W H and ZH processes and the three lepton channels with the combined result are at the level of 8 % in both cases. This compatibility is calculated performing a χ2 test, where fit re-sults are compared. The high signal strength associated with the 1-lepton, around two times SM expected, is correlated with the large excess of events above the expected background observed in Figure 8.4a atmbb = 125 GeV.
Since most of theW H signal is obtained from the 1-lepton channel, similar values are expected between their associated signal strengths. Instead, the ZH signal is obtained from 0- and 2-leptons analyses. Non-trivial correla-tions among these analyses explain the differences betweenZH and 0- and 2-lepton signal strengths which show a compatibility at the level of 15 %.
8.2. DIJET MASS ANALYSIS RESULTS
(a) (b)
(c)
Figure 8.3: The dijet-mass distribution observed in data (points with error bars) and expected (histograms) in the 0-lepton channel with the Medium and Tight b-tagging categories combined for (a) the 2-jet signal region in the 100< pVT < 120 GeV interval, (b) the 2-jet signal regions in the three intervals withpVT >120 GeV combined, and (c) the 3-jet signal regions in the three intervals withpVT >120 GeV combined. The background contributions after the global fit of the dijet-mass analysis are shown as filled histograms.
The Higgs boson signal (mH = 125 GeV) is shown as a filled histogram on top of the fitted backgrounds, as expected from the SM (indicated as µ = 1.0), and, unstacked as an unfilled histogram, scaled by the factor indicated in the legend. The dashed histogram shows the total background as expected from the pre-fit MC simulation. The entries in overflow are included in the last bin. The size of the combined statistical and systematic uncertainty on the sum of the signal and fitted background is indicated by the hatched band. The ratio of the data to the sum of the signal and fitted background is shown in the lower panel.
CHAPTER 8. RESULTS
(a) (b)
(c) (d)
Figure 8.4: The dijet-mass distribution observed in data (points with er-ror bars) and expected (histograms) with the Medium and Tightb-tagging categories combined and the three intervals with pVT > 120 GeV combined for (a) the 2-jet signal regions of the 1-lepton channel, (b) the 3-jet signal regions of the 1-lepton channel, (c) the 2-jet signal regions of the 2-lepton channel, and (d) the 3-jet signal regions of the 2-lepton channel. The back-ground contributions after the global fit of the dijet-mass analysis are shown as filled histograms. The Higgs boson signal (mH = 125 GeV) is shown as a filled histogram on top of the fitted backgrounds, as expected from the SM (indicated as µ= 1.0), and, unstacked as an unfilled histogram, scaled by the factor indicated in the legend. The dashed histogram shows the to-tal background as expected from the pre-fit MC simulation. The entries in overflow are included in the last bin. The size of the combined statistical and systematic uncertainty on the sum of the signal and fitted background is indicated by the hatched band. The ratio of the data to the sum of the signal and fitted background is shown in the lower panel.
8.2. DIJET MASS ANALYSIS RESULTS
[GeV]
mH
110 115 120 125 130 135 140
SMσ/σ 95% C.L. limit on
Figure 8.5: Observed (solid) and expected 95 % CL cross section upper lim-its, normalized to the SM Higgs boson production cross section, as a function ofmH for all channels using the 8 TeV dijet mass analysis dataset. The ex-pected upper limit is given for the background-only hypothesis (dashed) and with the injection of a SM Higgs boson signal at a mass of 125 GeV (dot-ted). The dark and light shaded bands represent the 1σ and 2σ ranges of the expectation in the absence of a signal.
[GeV]
mH
110 115 120 125 130 135 140
0Local p
Figure 8.6: Observed (solid) and expected p0 values as a function of mH
for all channels using the 8 TeV dijet mass analysis dataset. The expected p0 values are given for the background-only hypothesis in the presence of a SM Higgs boson: for the dashed curve the Higgs boson mass corresponds to each tested mass point in turn; for the dotted curve the Higgs boson mass is 125 GeV.
CHAPTER 8. RESULTS
Figure 8.7: The fitted values of the Higgs boson signal strength parameterµ formH = 125 GeV for the 0-, 1- and 2-lepton channels and the combination of the three channels with the 8 TeV dataset. The individual µ values for the lepton channels are obtained from a simultaneous fit with the signal strength for each of the lepton channels floating independently.
=125 GeV
Figure 8.8: The fitted values of the Higgs boson signal strength parameter µformH = 125 GeV for theW H andZH processes and the combination of theW H andZH processes with 8 TeV dataset. The individualµvalues for the (W/Z)H processes are obtained from a simultaneous fit with the signal strength for each of theW H and ZH processes floating independently.