The Bayes factor surface for searches for new physics

TL;DR

The paper introduces the Bayes factor surface as a new visualization tool for experimental searches in physics, providing a clear measure of evidence across parameters without prior dependence.

Contribution

It presents the Bayes factor surface as an innovative method for visualizing and interpreting evidence in physics experiments, independent of prior choices.

Findings

Effective visualization of evidence across parameters

Applicable to dark matter, cosmology, and collider physics

Enhances interpretation of experimental results

Abstract

The Bayes factor surface is a new way to present results from experimental searches for new physics. Searches are regularly expressed in terms of phenomenological parameters - such as the mass and cross-section of a weakly interacting massive particle. Bayes factor surfaces indicate the strength of evidence for or against models relative to the background only model in terms of the phenomenological parameters that they predict. They provide a clear and direct measure of evidence, may be easily reinterpreted, but do not depend on choices of prior or parameterization. We demonstrate the Bayes factor surface with examples from dark matter, cosmology, and collider physics.