A Practical Guide to Sensitivity Projections for Beyond the Standard Model Physics
Abstract
Statistics is a well-established discipline, yet its application can be confusing, particularly in fields where the language of statistics is unfamiliar and when making projections, where real data is not yet available. Both of these factors are commonplace in Beyond the Standard Model physics. In this work, we clarify the distinction between goodness-of-fit tests, which assess the compatibility of an observation with a single hypothesis, and hypothesis tests, which quantify how strongly some data prefer one hypothesis over another. We emphasise that for Beyond the Standard Model sensitivity projections, hypothesis testing should be used. We give detailed examples using both Poisson and multinomially distributed data, deriving exact p-values and discussing Monte Carlo and asymptotic approximations. We further demonstrate that for the Asimov data set all the test statistics discussed, for both goodness-of-fit tests and hypothesis tests, can be approximated by Pearson's χ2 statistic, and the difference in approaches comes down to the distribution of the test statistics. We also compare goodness-of-fit and hypothesis test p-values in two realistic examples. We (1) consider the reach of a muon collider on heavy spin-1 resonances and (2) provide updated projections for the potential of current and future gamma ray telescopes to probe new dark sectors via the observation of Hawking radiation from the explosion of a black hole. Finally, we discuss the CLs method, the significance and systematic errors.
Related
- Search for physics beyond the standard model in four and three top quark production events using proton-proton collisions at √s = 13 TeV
- A Practical Introduction to VQE: Methods, Challenges, and Applications in Physics and Chemistry
- CP violation in B(s)→ϕK decays: Standard Model and isospin-dependent New Physics
- Cosmology beyond standard sirens: cross-correlation of gravitational waves and neutral hydrogen intensity mapping
- Chiral Symmetries in the Standard Model of Particle Physics and its Extensions
BibTeX
@misc{acanfora2026practical,
title = {{A Practical Guide to Sensitivity Projections for Beyond the Standard Model Physics}},
author = {Francesca Acanfora and Michael J. Baker and Joaquim Iguaz Juan and Aidan Symons and Andrea Thamm},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.36030},
eprint = {2610.36030},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.36030},
}
RIS
TY - UNPB
TI - A Practical Guide to Sensitivity Projections for Beyond the Standard Model Physics
AU - Francesca Acanfora
AU - Michael J. Baker
AU - Joaquim Iguaz Juan
AU - Aidan Symons
AU - Andrea Thamm
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.36030
UR - https://arxiv.org/abs/2610.36030
AB - Statistics is a well-established discipline, yet its application can be confusing, particularly in fields where the language of statistics is unfamiliar and when making projections, where real data is not yet available. Both of these factors are commonplace in Beyond the Standard Model physics. In this work, we clarify the distinction between goodness-of-fit tests, which assess the compatibility of an observation with a single hypothesis, and hypothesis tests, which quantify how strongly some data prefer one hypothesis over another. We emphasise that for Beyond the Standard Model sensitivity projections, hypothesis testing should be used. We give detailed examples using both Poisson and multinomially distributed data, deriving exact p-values and discussing Monte Carlo and asymptotic approximations. We further demonstrate that for the Asimov data set all the test statistics discussed, for both goodness-of-fit tests and hypothesis tests, can be approximated by Pearson's χ^2 statistic, and the difference in approaches comes down to the distribution of the test statistics. We also compare goodness-of-fit and hypothesis test p-values in two realistic examples. We (1) consider the reach of a muon collider on heavy spin-1 resonances and (2) provide updated projections for the potential of current and future gamma ray telescopes to probe new dark sectors via the observation of Hawking radiation from the explosion of a black hole. Finally, we discuss the CLs method, the significance and systematic errors.
ER -