Period of duration of course
PhD students, as well as 4th and 5th year undergraduates
This course deals with statistical data analysis, with an application towards astronomical and cosmological observations. Starting from Bayes' theorem, we show how it can be used for data-driven inference, including analytical, Monte Carlo, as well as machine learning implementations. Topics to be covered include: introduction to probability, maxlimum likelihood estimation, Bayesian vs classical frequentist statistics, parameter inference (analytic and Monte Carlo methods), model selection, likelihood free inference, data compression, introduction to neural networks and their applications in inference. Examples will be given throughout, with a focus towards topical problems in astrophysics and cosmology.
The students should come away with an understanding of Bayesian inference, including some practical applications.
materials will be provided during the course