Bayesian analysis of EFTs with Jupyter notebooks
Abstract
Uncertainty quantification (UQ) is an essential part of applying effective field theories (EFT) to low-energy nuclear physics. A Bayesian statistical framework is particularly well suited for this task, as EFT expectations regarding naturalness and truncation errors can be encoded through prior probability distribution functions (PDFs). The specification of priors means that all theoretical assumptions are explicit in the calculation of the posterior PDFs, making such an analysis reproducible. The BUQEYE collaboration (``Bayesian Uncertainty Quantification: Errors for Your EFT'') has the overall goal of full UQ and associated diagnostics for EFT predictions using Bayesian statistics. The BUQEYE website https://buqeye.github.io/ has freely available Python code, Jupyter notebooks, and cheatsheets to make reproducing and extending our results easy. In this talk we provide a guide to this material.
Supported in part by the NSF and the DOE.- Publication:
-
APS Division of Nuclear Physics Meeting Abstracts
- Pub Date:
- 2020
- Bibcode:
- 2020APS..DNP.SJ001F