Constraints on anomalous HVV interactions using H -> 4l decays


pdf,png
Table of contents:

The Physics Analysis Summary for this study is available at CDS as CMS-PAS-HIG-14-014.

Abstract

In this document we present cons the study of anomalous Higgs interactions with pairs of neutral electroweak gauge bosons performed using the H → 4l decay mode. The results are based on the 2011 and 2012 data recorded with the CMS detector at the LHC, corresponding to an integrated luminosity of 5.1 f b-1 at center-of-mass energy of √s = 7 TeV and 19.7 f b-1 at √s = 8 TeV. Under the assumption that the resonance is a spin-zero boson, we probe for the presence of anomalous effects in Higgs interactions with ZZ, Zγ* and γ*γ* boson pairs in the 4l final state, with l = e, μ. We find that our observations are consistent with the expectations for the standard model Higgs boson and we put the constrains on these anomalous interactions. We also perform tests of alternative spin-parity hypotheses to cover spin-one and spin-two resonances which were not tested in previous publications. The tested spin-two boson hypotheses are excluded at a 95% confidence level or higher, while any mixture of a vector and a pseudo-vector state is excluded at a 99% confidence level or higher. We also extend the study by including the information from the H → WW → lνlν decay mode. This additional information puts tighter constraints on the spin-zero anomalous interactions both under the assumption of custodial symmetry and without this assumption. Furthermore, this additional information allowed us to exclude all tested spin-one and spin-two boson hypotheses at a 99.9% confidence level or higher.

Results obtained using the H → 4l decays

General (event yields, list of discriminants, systematics)

Figure Description
The number of estimated background and signal events, and number of observed candidates, after final inclusive selection in the fullmeasurement range 105.6 < m4l < 140.6 GeV. Signal and ZZ background are estimated from Monte Carlo simulation, while Z+X background is estimated from data.

List of kinematic discriminants used in these analyses.

Summary of the systematics used in this analysis. The implementation of each systematic may be different in the two analysis methods but all are considered in both frameworks. *The "production dependency" systematic is only applied when the hypothesis testing is used in a production-independent scenario.

Spin-0 anomalous couplings

Observables and kinematic discriminants

Figure Description
Distributions of kinematic observables in data, as well as the expectations for the SM background, the Higgs boson signal and indicated alternative spin-zero scenarios (signal resonance at mH = 125.6 GeV). From left to right: m4l, m1, m2. All distributions, with exception of the m4l, are shown with the requirement 121.5 < m4l < 130.5 GeV to enhance the signal purity.

Distributions of kinematic observables in data, as well as the expectations for the SM background, the Higgs boson signal and indicated alternative spin-zero scenarios (signal resonance at mH = 125.6 GeV). From left to right: cos θ1, cos θ2, cos θ. All distributions, with exception of the m4l, are shown with the requirement 121.5 < m4l < 130.5 GeV to enhance the signal purity.



Distributions of kinematic observables in data, as well as the expectations for the SM background, the Higgs boson signal and indicated alternative spin-zero scenarios (signal resonance at mH = 125.6 GeV). From left to right: Φ, Φ1. All distributions, with exception of the m4l, are shown with the requirement 121.5 < m4l < 130.5 GeV to enhance the signal purity.

Figure Description
Distributions of kinematic discriminants in data, as well as the expectations for the SM background, the Higgs boson signal and indicated alternative spin-zero scenarios (signal resonance at mH = 125.6 GeV). From left to right: Dbkg, D0−, DCP. All distributions, with exception of the Dbkg, are shown with the requirement Dbkg > 0.5 to enhance signal purity. Discriminants DCP and Dint provide the sensitivity to the interference between the (SM, 0) and (SM, 0h+ ) pairs of terms in the ZZ amplitudes, respectively.

Distributions of kinematic discriminants in data, as well as the expectations for the SM background, the Higgs boson signal and indicated alternative spin-zero scenarios (signal resonance at mH = 125.6 GeV). From left to right: D0h+, Dint, DΛ1. All distributions, with exception of the Dbkg, are shown with the requirement Dbkg > 0.5 to enhance signal purity. Discriminants DCP and Dint provide the sensitivity to the interference between the (SM, 0) and (SM, 0h+ ) pairs of terms in the ZZ amplitudes, respectively.

Probing a single spin-zero anomalous coupling

Figure Description
Expected and observed likelihood scans for fa2(left) and fa3(right) obtained using the kinematic discriminant method (KD, black) and multidimensional distribution method (MD, red). The likelihoods are computed in the two methods assuming the a2/a1 and a3/a1 coupling ratios are real.

Figure Description
Expected and observed likelihood scans for effective fractions fΛ1 from the kinematic discriminant method. Left column shows the results where the Λ1 amplitude is constrained to be real, and all other amplitudes are fixed to the SM predictions. The right column shows the results where the phase of the amplitude, as well as additional ZZ amplitudes are profiled.

Expected and observed likelihood scans for effective fractions fa2 from the kinematic discriminant method. Left column shows the results where the a2 amplitude is constrained to be real, and all other amplitudes are fixed to the SM predictions. The right column shows the results where the phase of the amplitude, as well as additional ZZ amplitudes are profiled.

Expected and observed likelihood scans for effective fractions fa3 from the kinematic discriminant method. Left column shows the results where the a3 amplitude is constrained to be real, and all other amplitudes are fixed to the SM predictions. The right column shows the results where the phase of the amplitude, as well as additional ZZ amplitudes are profiled. Result for fa3 with phase φa3 profiled in the top right plot has been obtained in the analysis in Ref. [10] in the PAS.

Figure Description
Expected and observed likelihood scans for fa2 and fa3 fractions from the kinematic discriminant method. The corresponding amplitudes are constrained to be real, while all other amplitudes are fixed to their SM predictions.

Expected and observed likelihood scans fa2γγ and fa3γγ fractions from the kinematic discriminant method. The corresponding amplitudes are constrained to be real, while the all other amplitudes are fixed to their SM predictions.

Figure Description
Summary of allowed one-dimensional 95% CL intervals from kinematic discriminant method and multidimensional distribution method under the assumption that all the coupling ratios are real (φai=0 or π). For fΛ1, and fa2γγ we also report allowed intervals extracted separately for positive-only (φai=0) or negative-only (φai=π) case because the minimum appears >1σ away from SM in data. Unless otherwise specified the other amplitudes are assumed to be the SM prediction. Symbol `--' denotes that the current analysis still allows the full region [0.00,1.00].

Summary of allowed one-dimensional 95% CL intervals from kinematic discriminant method under the assumption that the couplings can be complex. Unless otherwise specified the other amplitudes are assumed to be the SM prediction. Result for fa3 with φa3 profiled has been obtained in the analysis presented in Ref.[10] and is quoted for completeness. Symbol `--' denotes that the current analysis still allows the full region [0.00,1.00].

Summary to cross section ratios for a 125.6 GeV Higgs boson and the scale Λ0 used to convert the effective fractions fai into the equivalent parameters of interest. The σi is the cross section of the process corresponding to ai = 1, aj≠i = 0 in the H to 2e2μ final state. For the Zγ**γ*) terms all cross sections σ1, σ2, σ3, σ2γγ and σ3γγ are given for m1 ≥ m2 ≥ 4GeV.

Summary of allowed one-dimensional 95% CL intervals on the ratios of anomalous couplings with respect to the SM coupling a1, obtained using kinematic discriminant method and multidimensional distribution method and assuming the ratios of couplings are real. In case of the Zγ* and γ*γ* we also interpret the allowed intervals in terms of the ratio of corresponding cross-sections with respect to the SM prediction. Unless otherwise specified the other amplitudes are assumed to be the SM prediction (a1 = 2, a2=0.0035, and a2γγ=-0.0040).

Probing a pair of spin-zero anomalous couplings

Figure Description


The observed 2D likelihood scan for fa2 vs fa3 obtained using the kinematic discriminant method (black) and multidimensional distribution method (red). The likelihoods are computed in the two methods assuming the a2/a1 and a3/a1 coupling ratios are real.

Figure Description
Expected likelihood scans for pairs of effective fractions fΛ1 vs fa2. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Observed likelihood scans for pairs of effective fractions fΛ1 vs fa2. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Figure Description
Expected likelihood scans for pairs of effective fractions fΛ1 vs fa3. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Observed likelihood scans for pairs of effective fractions fΛ1 vs fa3. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Figure Description
Expected likelihood scans for pairs of effective fractions fa2 vs fa3. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Observed likelihood scans for pairs of effective fractions fa2 vs fa3. In plots in the left column amplitudes are constrained to be real, and all other amplitudes are fixed to the SM. In plots in the right column the phases of amplitudes are profiled. Results are obtained using the kinematic discriminant method.

Figure Description
Likelihood scans for (fa2, φa2). expected results are on the left, while observed are shown on the right. Results are obtained using the kinematic discriminant method.

Figure Description
Likelihood scans for (fΛ1, φΛ1). expected results are on the left, while observed are shown on the right. Results are obtained using the kinematic discriminant method.

Exotic models

Spin-one results

Figure Description
Distribution of a test-statistic q = −2 ln(LJP /L0+ ) of an example spin-one hypothesis with mixture fb2=0.8 tested against the SM Higgs boson hypothesis, for the production dependent scenario. Distributions for the SM Higgs boson are represented by the yellow histogram and for the alternative JP hypotheses by the blue histogram (left). Expected and observed distribution of −2∆ ln(L) as a function of f (JP) for 1+, for the qq ̄ → X → ZZ case (right).

Figure Description
The expected and observed distributions of median test-statistic q for alternative mixed spin-one hypotheses, as function of fb2 . The green and blue band represents the 1σ and 2σ around the median expected value for the SM Higgs boson hypothesis. Left and right plots represent results for the cases of production independent and production dependent analysis, respectively.

Figure Description
Summary of expected and observed constraints on the measurements of the non-interfering fractions f (JP) for the points used in the scan of the fb2 fraction. In the case of production independent scenarios the f (JP) measurement is performed as using the efficiency of qq → X.

Figure Description
List of models used in the analysis of the spin and parity hypotheses corresponding to the spin-one pure and mixture states of the type noted. The expected separation is quoted for two scenarios, where the signal strength for each hypothesis is predetermined from the fit to data and where events are generated with SM expectations for the signal cross section (μ=1). The observed separation quotes consistency of the observation with the 0+ model or JP model and corresponds to the scenario where the signal strength is floated in the fit to data. We also quote the CLs value for the JP model and the 95% CL and best fit of the f(JP) fraction. In the case of production independent scenarios the f(JP) measurement is performed as using the efficiency of qq to X.

Spin-two results

Figure Description
Distribution of the test statistic q = −2ln(LJP /L0+ ) of the hypothesis any → 2-h10 tested against the SM Higgs boson hypothesis for mH = 125.6 GeV. Distributions for the SM Higgs boson are represented by the yellow histogram, and those for the alternative JP hypothesis are represented by the blue histogram. The red arrow indicates the observed value of test-statistics (left). Expected and observed distribution of −2∆ ln L for the gg → 2-h10 model as a function of on the fractional presence f ( J P ) of J P model as a state nearly degenerate with the 0+ state. (right).

Figure Description
Summary of the expected and observed values for the test-statistic q distributions for the twelve alternative spin-two hypotheses tested with respect to the SM Higgs boson. The orange (blue) bands represent the 1σ, 2σ, and 3σ around the median expected value for the SM Higgs boson hypothesis (alternative hypothesis). The black point represents the observed value.

Figure Description
Summary expected and observed constraints on the non-interfering fraction measurements for the fraction. In the case of production independent scenarios the f ( J P ) measurement is performed as using the efficiency of qq → X.

Figure Description
List of models used in the analysis of the spin-two hypotheses corresponding to the pure states of the type noted. The expected separation is quoted for two scenarios, where the signal strength for each hypothesis is predetermined from the fit to data and where events are generated with SM expectations for the signal cross section (μ=1). The observed separation quotes consistency of the observation with the 0+ model or JP model and corresponds to the scenario where the signal strength is floated in the fit to data. We also quote the CLs value for the JP model and the 95% CL and best fit of the f(JP) fraction. In the case of production independent scenarios the f(JP) measurement is performed as using the efficiency of gg to X.

Validation of hypothesis testing in Z → 4l decays in data

Figure Description
Distribution of the test statistic q = −2ln(L0+ /LZ4l ) of the standard model hypothesis Z0→4l tested against the alternative exotic Higgs hypothesis 0+ → 4l, where mass and width of the exotic particle match those of the SM Z0 boson (m0+ = 91.2 GeV, width 2.5 GeV). Distributions for the SM Z0 boson are represented by the yellow histogram, and those for the alternative hypothesis are represented by the blue histogram. The red arrow indicates the observed value of test-statistics (left). Average expected and observed distribution of −2∆ ln(L) for the fractional presence f(JP) of the exotic alternative Higgs model as a state nearly degenerate with the SM Z0 boson, as measured in the 4l decays in the 91.2 GeV peak in data (right).

Results obtained using the H → 4l and H → WW → lνlν decays

Spin-0 anomalous couplings

Figure Description
Observed conditional scans of fai given Rai value. The fΛ1 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right) . This combination uses the kinematic discriminant method from H → ZZ.

Observed conditional scans of fai given Rai value. The fa2 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right) . This combination uses the kinematic discriminant method from H → ZZ.

Observed conditional scans of fai given Rai value. The fa3 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right) . This combination uses the kinematic discriminant method from H → ZZ.

Expected conditional scans of fai given Rai value. The fΛ1 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right). This combination uses the kinematic discriminant method from H → ZZ.

Expected conditional scans of fai given Rai value. The fa2 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right). This combination uses the kinematic discriminant method from H → ZZ.

Expected conditional scans of fai given Rai value. The fa3 results are presented with the assumption of custodial symmetry aWW = aZZ (left), and without this assumption (right). This combination uses the kinematic discriminant method from H → ZZ.

Figure Description
The combined result of H → WW and H → ZZ (kinematic discriminant method) for fΛ1, fa2, fa3 measurements (left to right). Two individual measurements are shown in each channel, where the H → WW measurement is related for the case Rai = 0.5(rai = 1). Two scenarios of combination of measurements are presented: using custodial symmetry aWW = aZZ (red) and without such a constraint (magenta), shown for Rai = 0.5(rai = 1) in both case.

Exotic models

Figure Description
Distributions of a test-statistic q = −2 ln(LJP /L0+ ) of the hypotheses of the spin-one boson with positive parity (left) and negative parity (right), tested against the SM Higgs boson hypothesis, for the qq ̄ production scenario. Distributions for the SM Higgs boson are represented by the yellow histogram and for the alternative JP hypotheses by the blue histogram. The red arrow indicates the observed value of test-statistics.

Figure Description
Distributions of a test-statistic q = −2 ln(LJP /L0+ ) of the example spin-two hypotheses gg → 2h6+ (left) and gg → 2h10- (right), tested against the SM Higgs boson hypothesis for mH = 125.6 GeV. Distributions for the SM Higgs boson are represented by the yellow his- togram and for the alternative JP hypotheses by the blue histogram. The red arrow indicates the observed value of test-statistics.

Figure Description
Summary of the expected and observed values for the test-statistic q distributions for the alternative spin-one and spin-two hypotheses tested with respect to the SM Higgs boson, based on the combined analysis in H → ZZ → 4l and H → WW → lνlν decay channels. The orange (blue) bands represent the 1σ, 2σ, and 3σ around the median expected value for the SM Higgs boson hypothesis (alternative hypothesis). The black point represents the observed value.

Figure Description
Results of the study of the spin-one and spin-two hypotheses corresponding to the pure states of the type noted, based on the combined analysis in H to ZZ to 4l and H to WW to lνlν decay channels. The expected separation is quoted for two scenarios, where the signal strength for each hypothesis is predetermined from the fit to data (fitted independently in the HWW and HZZ channels), and where events are generated with SM expectations for the signal cross section (μ=1). The observed separation quotes consistency of the observation with the 0+ model or JP model and corresponds to the scenario where the signal strength is floated in the fit to data. Table does not quote the CLs value for the JP model because all values are <0.1%.
Edit | Attach | Watch | Print version | History: r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r2 - 2014-07-03 - PredragMilenovic
 
    • 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 & 2008-2021 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