2022
DOI: 10.1088/1361-6633/ac60ac
|View full text |Cite
|
Sign up to set email alerts
|

Simple and statistically sound recommendations for analysing physical theories

Abstract: Physical theories that depend on many parameters or are tested against data from many different experiments pose unique challenges to statistical inference. Many models in particle physics, astrophysics and cosmology fall into one or both of these categories. These issues are often sidestepped with statistically unsound ad hoc methods, involving intersection of parameter intervals estimated by multiple experiments, and random or grid sampling of model parameters. Whilst these methods are easy to apply, they ex… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
2

Relationship

2
7

Authors

Journals

citations
Cited by 15 publications
(6 citation statements)
references
References 96 publications
0
6
0
Order By: Relevance
“…In this section we perform a global analysis of our model with three generations of the Φ 2 leptoquark. For this we construct full likelihood functions for the LQ parameters whenever possible [245]. Implementing the formulas of section 3, we use for the numerical analysis the software packages flavio [179] and smelli [196].…”
Section: Phenomenological Analysismentioning
confidence: 99%
“…In this section we perform a global analysis of our model with three generations of the Φ 2 leptoquark. For this we construct full likelihood functions for the LQ parameters whenever possible [245]. Implementing the formulas of section 3, we use for the numerical analysis the software packages flavio [179] and smelli [196].…”
Section: Phenomenological Analysismentioning
confidence: 99%
“…The 90% CL spin-independent limits of Xenon1T for DM direct detection and 95% CL limits from HiggsBounds are also applied. As a result, the overall confidence level of the displayed points are stricter than 2σ level [73]. The horizontal and vertical solid, dashed and dot-dashed lines indicate the current central value, 1σ limits and 2σ limits of the corresponding observable.…”
Section: Resultsmentioning
confidence: 94%
“…In particular, we will combine observational data with theoretical constraints-from perturbative unitarity-and it is clearly unreasonable to combine them statistically. In our applications, in order to build a new class of search algorithm capable of using the latest versions of the relevant codes we shall take our classification based on simple criteria and consider a point to be "bad" if it fails the test for one observable; this has been argued to be a poor choice for global scans [57] but is at least unambiguous and easy to implement, and should be a good approximation for the cases we are interested in.…”
Section: Active Learning With a Neural Networkmentioning
confidence: 99%