Practical Statistics for Particle Physics

Abstract

This is the write-up of a set of lectures given at the Asia Europe Pacific School of High Energy Physics in Quy Nhon, Vietnam in September 2018, to an audience of PhD students in all branches of particle physics They cover the different meanings of 'probability', particularly frequentist and Bayesian, the binomial, Poisson and Gaussian distributions, hypothesis testing, estimation, errors (including asymmetric and systematic errors) and goodness of fit. Several different methods used in setting upper limits are explained, followed by a discussion on why 5 sigma are conventionally required for a 'discovery'.

References in corpus (3)

• A Unified Approach to the Classical Statistical Analysis of Small Signals
• Measurement of CP-violating asymmetries in B0 decays to CP eigenstates
• A note on Delta ln L = -1/2 Errors

Cited by in corpus (4)

• The IceCube high-energy starting event sample: Description and flux characterization with 7.5 years of data
• Application of the Most Frequent Value Method for ${}^{39}$Ar Half-Life Determination
• First Search for High-Energy Neutrino Emission from Galaxy Mergers
• Fast estimation of the look-elsewhere effect using Gaussian random fields