Speaker
Description
Properly constructed effective field theories (EFTs) hold the promise of model independence and order-by-order convergence of observable calculations.
EFTs have free parameters, the so-called low-energy constants (LECs), that often must be fixed using low-energy data relative to the breakdown scale of the EFT. We have developed a Bayesian framework for EFT parameter estimation that uses priors to encode information about LEC naturalness and EFT convergence. This allows us to avoid common parameter estimation problems and makes all assumptions explicit. Bayesian model selection is used to guide the parameter estimation by testing our assumptions about naturalness and making quantitative estimates of how many parameters can actually be constrained by the data available. Bayesian model checking is also used to verify whether EFT predictions converge as expected and to estimate the intrinsic breakdown of the theory.
