The correlation matrix of Higgs rates at the LHC

Arbey, Alexandre; Moreau, Grégory; Mahmoudi, Farvah; Fichet, Sylvain

doi:10.1007/JHEP11(2016)097

Article
Report number	arXiv:1606.00455 ; CERN-TH-2016-130 ; CERN-TH-2016-130
Title	The correlation matrix of Higgs rates at the LHC
Author(s)	Arbey, Alexandre (Lyon U. ; CERN) ; Fichet, Sylvain (ICTP-SAIFR, Sao Paulo) ; Mahmoudi, Farvah (Lyon U. ; CERN) ; Moreau, Grégory (Orsay, LPT)
Publication	2016-11-17
Imprint	01 Jun 2016
Number of pages	24
Note	Comments: 24 pages, 1 figure. Complete covariance matrix is available in C, Fortran, Mathematica, PDF, TeX and text formats in ancillary files 25 pages, 1 figure. Complete covariance matrix is available in C, Fortran, Mathematica, PDF, TeX and text formats in ancillary files
In:	JHEP 11 (2016) 097
DOI	10.1007/JHEP11(2016)097
Subject category	Particle Physics - Phenomenology
Abstract	The imperfect knowledge of the Higgs boson LHC cross sections and decay rates constitutes a critical systematic uncertainty in the study of the Higgs boson properties. We show that the full covariance matrix between the Higgs rates can be determined from the most elementary sources of uncertainty by a direct application of probability theory. We evaluate the error magnitudes and full correlation matrix on the set of Higgs cross sections and partial decay widths at $\sqrt{s}=7$, $8$, $13$ and $14$ TeV, which are provided in ancillary files. The impact of this correlation matrix on the global fits is illustrated with the latest $7$+$8$ TeV Higgs dataset.
Copyright/License	arXiv nonexclusive-distrib. 1.0 preprint: © 2016-2025 CERN (License: CC-BY-4.0) publication: © 2016-2025 The Author(s) (License: CC-BY-4.0), sponsored by SCOAP³

$Best-fit regions in the $c_V-c_f$ plane obtained from Bayesian marginalisation. The $68\%$, $95\%$ and $99\%$ credible regions when theoretical uncertainties are taken into account are represented respectively by the green, yellow and grey domains. The dashed contours illustrate the case without theoretical uncertainties. The SM prediction is shown by a red point. The two extreme cases for renormalisation and factorisation scale correlations, leading to the two plots, are described in Eqs.~\eqref{eq:Corri}, \eqref{eq:Corrii}.$ $Best-fit regions in the $c_V-c_f$ plane obtained from Bayesian marginalisation. The $68\%$, $95\%$ and $99\%$ credible regions when theoretical uncertainties are taken into account are represented respectively by the green, yellow and grey domains. The dashed contours illustrate the case without theoretical uncertainties. The SM prediction is shown by a red point. The two extreme cases for renormalisation and factorisation scale correlations, leading to the two plots, are described in Eqs.~\eqref{eq:Corri}, \eqref{eq:Corrii}.$