Extreme Value Theory: Point Processes

This lecture covers the application of extreme value theory to point processes, focusing on the estimation of extreme events from equally-spaced time series. The GEV limit is assumed for maxima, leading to the formation of two-dimensional binomial processes. The lecture delves into the use of Kallenberg's theorem for convergence, the construction of binomial processes, and the application of Poisson processes. The instructor discusses the estimation of parameters, the Poisson process limit, and the implications of Kallenberg's conditions. The lecture concludes with the interpretation of results and the graphical approach to threshold selection.