Speaker
Description
The field of neutrino oscillation study is full of unique and insightful experiments, and global fits can be conducted to study their results in a unified and coherent approach, exploiting the strengths of the different experiments. For the success of a global study, factors such as experiment modelling, parameter space exploration, and statistical interpretation are of vital importance.
In this work, we present preliminary results from the first three-flavour neutrino global fit performed with the Global and Modular BSM Inference Tool (GAMBIT). GAMBIT is an open-source global fitting software package for studying generic particle and astronomical physics models. Its modular design allows easy implementation of likelihood functions and models. The built-in scanners also provide robust and efficient statistical sampling techniques.
Our neutrino global fit includes results from eight neutrino oscillation experiments of different types, including solar, reactor, atmospheric, and long-baseline accelerator. The fit also uses only publicly accessible experiment data and information, adhering to the open-source policy. In the fit, each experiment is represented by a set of likelihood functions. Realistic and physics-motivated systematic models along with sets of nuisance parameters are introduced to account for systematic uncertainties for the detector effects and the neutrino fluxes, to name a few. For a given combination of neutrino oscillation parameters and nuisance parameters, a combined likelihood can be calculated. A self-adaptive differential evolution sampling algorithm is utilised to explore the vast parameter space and search for the best-fit point. Rigorous and modern statistical methods are adopted to interpret the sampling result, maximising the accuracy of the global fit.
| Poster prize | Yes |
|---|---|
| Given name | Chien |
| Surname | Lin |
| First affiliation | Imperial College London |
| Institutional email | c.lin20@imperial.ac.uk |
| Gender | Male |
| Collaboration (if any) | The GAMBIT Collaboration |
