Montréal Joint High Energy Physics Seminar

A Bayesian Revolution in Spectral Analysis

Phil Gregory

UBC

Science is all about identifying and understanding organized structures or patterns in nature. Periodic patterns have proven especially important. My talk will focus on recent advances in our ability to detect and quantify periodic phenomena that are based on Bayesian probability theory, a mathematical framework that encompasses both inductive and deductive logic. Starting with new insights concerning the Fourier power spectrum, examples will be given from a variety of fields in physics and astronomy.