Robust Determination of the Higgs Couplings: Power to the Data

TL;DR

The paper develops a model-independent EFT framework to study indirect new-physics effects in Higgs phenomenology, assuming a linearly realized SU(2)_L x U(1)_Y with dimension-6 operators. It advocates a data-driven operator basis, reducing to a nine-operator set that directly affects Higgs couplings to gluons, electroweak gauge bosons, and fermions, and performs a global chi-square fit to Higgs signal strengths, TGV measurements, and EW precision data. The results show SM predictions for individual Higgs couplings are consistent within 68% CL, with no strong evidence for large deviations; including fermionic operators can introduce degeneracies, but unitarity considerations imply new physics scales above ~2 TeV. Overall, the work demonstrates that current data tightly constrain Higgs couplings in a model-independent way and outlines how future data will sharpen these bounds.

Abstract

We study the indirect effects of new physics on the phenomenology of the recently discovered "Higgs-like" particle. In a model independent framework these effects can be parametrized in terms of an effective Lagrangian at the electroweak scale. In a theory in which the SU(2)_L x U(1)_Y gauge symmetry is linearly realized they appear at lowest order as dimension--six operators, containing all the SM fields including the light scalar doublet, with unknown coefficients. We discuss the choice of operator basis which allows us to make better use of all the available data on the new state, triple gauge boson vertex and electroweak precision tests, to determine the coefficients of the new operators. We illustrate our present knowledge of those by performing a global fit to the existing data which allows simultaneous determination of the eight relevant parameters quantifying the Higgs couplings to gluons, electroweak gauge bosons, bottom quarks, and tau leptons. We find that for all scenarios considered the standard model predictions for each individual Higgs coupling and observable are within the corresponding 68% CL allowed range. We finish by commenting on the implications of the results for unitarity of processes at higher energies. Note added: The analysis has been updated with all the public data available by October 2013. Updates of this analysis are provided at http://hep.if.usp.br/Higgs as well as new versions of this manuscript.

Figures (6)

• Figure 1: $\Delta\chi^2$ dependence on the fit parameters when we consider all Higgs collider (ATLAS, CMS and Tevatron) data (solid red line), Higgs collider and TGV data (dashed purple line) and Higgs collider, TGV and EWP data (dotted blue line). The rows depict the $\Delta\chi^2$ dependence with respect to the fit parameter shown on the left of the row with the anomalous couplings $f/\Lambda^2$ given in TeV$^{-2}$. In the first column we use $f_g$, $f_{WW}$, $f_{BB}$, $f_W$, $f_B$, and $f_{\Phi,2}$ as fit parameters with $f_{\rm bot} = f_\tau =0$. In the second column the fitting parameters are $f_g$, $f_{WW}=-f_{BB}$, $f_W$, $f_B$, $f_{\Phi,2}$, and $f_{\rm bot}$ with $f_\tau =0$. In the panels of the right column we fit the data in terms of $f_g$, $f_{WW}=-f_{BB}$, $f_W$, $f_B$, $f_{\Phi,2}$, $f_{\rm bot}$, and $f_\tau$.
• Figure 2: Chi--square dependence on the Higgs branching ratios (left panels) and production cross sections (right panels) when we consider all Higgs collider and TGV data. In the upper panels we used $f_g$, $f_{WW}$, $f_{BB}$, $f_W$, $f_B$, and $f_{\Phi,2}$ as fitting parameters with $f_{\rm bot} = f_\tau =0$, while in the middle panels the fitting parameters are $f_g$, $f_{WW}=-f_{BB}$, $f_W$, $f_B$, $f_{\Phi,2}$, and $f_{\rm bot}$ with $f_\tau =0$. In the lower row we parametrize the data in terms of $f_g$, $f_{WW}=-f_{BB}$, $f_W$, $f_B$, $f_{\Phi,2}$, $f_{\rm bot}$, and $f_\tau$. The dependence of $\Delta \chi^2$ on the branching ratio to the fermions not considered in the analysis arises from the effect of the other parameters in the total decay width.
• Figure 3: We display the 95% and 99% CL allowed regions in the plane $f_{WW} \times f_{BB}$ when we fit the Higgs collider data varying $f_g$, $f_{WW}$, $f_{BB}$, $f_W$, $f_B$, and $f_{\Phi,2}$ The star stands for the global minima and we marginalized over the undisplayed parameters.
• Figure 4: We present the 68%, 90%, 95%, and 99% CL allowed regions in the plane $f_g \times f_{\Phi,2}$ when we fit the Higgs collider and TGV data varying $f_g$, $f_{WW}$, $f_{BB}$, $f_W$, $f_B$, and $f_{\Phi,2}$. The stars stand for the global minima and we marginalized over the undisplayed parameters.
• Figure 5: In the left (right) panel we present the 68%, 90%, 95%, and 99% CL allowed regions in the plane $\sigma_{gg}^{\rm ano}/ \sigma_{gg}^{\rm SM} \times \hbox{Br}(h\to \gamma\gamma)^{\rm ano}/ \hbox{Br}(h\to \gamma\gamma)^{\rm SM}$ when we fit the Higgs collider and TGV data varying $f_g$, $f_{WW}$, $f_{BB}$, $f_W$, $f_B$, and $f_{\Phi,2}$ ($f_g$, $f_{WW}=-f_{BB}$, $f_W$, $f_B$, $f_{\Phi,2}$, and $f_{\rm bot}$). The stars stand for the global minima and we marginalized over the undisplayed parameters.