Handbook of Statistics of Extremes

Statistics of extremes is a prominent field of research concerned with modeling the risk of occurrence of extreme events, that is, low-probability-high-impact events such as a stock market crash, hurricanes, heatwaves, and widespread flooding.

The Handbook of Statistics of Extremes covers statistical models for univariate, multivariate, and spatio-temporal extreme values. Written by leading experts from around the world, it serves as a key reference for statisticians and data scientists, as well as for professionals working in risk modeling—such as geophysical and climate scientists, financial analysts, and health clinicians and neuroscientists—and as a valuable resource for practitioners and graduate students who wish to deepen their understanding of the statistical modeling of extreme events.

Key Features:

· Presents frequentist and Bayesian methods, as well as AI-based techniques for extreme value analysis.

· Details how to model the frequency, magnitude, and spatio-temporal dependence of extreme events, and how to extrapolate into the tails of a distribution beyond observed data.

· Provides code, data, and other additional materials available here: https://extremestats.github.io/Handbook/.