PHYS240

A survey of statistical and machine learning techniques as applied in modern physics research, with extensive applications to real data. Some of the topics covered include Bayesian and frequentist approaches to statistics; formulating and integrating likelihood functions; confidence intervals with and without assumptions of Gaussianity; Markov Chain Monte Carlo; principal component analysis and dimensionality reduction; Gaussian process regression; and time series analysis. Assignments will make use of published research and data sets, and will require the application of analysis techniques covered in class.

Vector calculus, linear algebra, some statistics and probability, some upper division physics, programming in python.