Table of Contents
Fetching ...

The gluon-fusion uncertainty in Higgs coupling extractions

Charalampos Anastasiou, Kirill Melnikov, Frank Petriello

Abstract

We point out that the QCD corrections to the gluon-fusion Higgs boson production cross section at the LHC are very similar to the corrections to the Higgs decay rate into two gluons. Consequently, the ratio of these two quantities has a theoretical uncertainty smaller than the uncertainty in the cross section alone by a factor of two. We note that since this ratio is the theoretical input to analyses of Higgs coupling extractions at the LHC, the reduced uncertainty should be used; in previous studies, the full cross section uncertainty was employed.

The gluon-fusion uncertainty in Higgs coupling extractions

Abstract

We point out that the QCD corrections to the gluon-fusion Higgs boson production cross section at the LHC are very similar to the corrections to the Higgs decay rate into two gluons. Consequently, the ratio of these two quantities has a theoretical uncertainty smaller than the uncertainty in the cross section alone by a factor of two. We note that since this ratio is the theoretical input to analyses of Higgs coupling extractions at the LHC, the reduced uncertainty should be used; in previous studies, the full cross section uncertainty was employed.

Paper Structure

This paper contains 3 equations, 3 figures.

Figures (3)

  • Figure 1: $\mu_R$ scale dependence for the Higgs production cross section at the LHC, as a function of $m_H$. $\mu_R$ is varied between $m_h/2 \leq \mu_R \leq 2m_h$, while $\mu_F = m_H$. The LO, NLO, and NNLO distributions are shown.
  • Figure 2: $\mu_R$ scale dependence for the $\sigma / \Gamma$ ratio at the LHC, as a function of $m_H$. $\mu_R$ is varied between $m_h/2 \leq \mu_R \leq 2m_h$, while $\mu_F = m_H$. The LO, NLO, and NNLO distributions are shown.
  • Figure 3: Theoretical error on $\sigma$ and $\sigma/\Gamma$ as a function of Higgs mass. The LO, NLO, and NNLO uncertainties are shown. The green and red lines at the bottom denote the theoretical errors at LO and NLO, respectively. As explained in the text, only at NNLO is a reliable error estimate dependence obtained.