Miguel de Carvalho is Reader in Statistics at the University of Edinburgh, UK. He is the current President of the Portuguese Statistical Society, and the former Director of the Centre for Statistics of the University of Edinburgh. Miguel has been an Associate Editor for a variety of top-tier journals in the field of Mathematical Statistics, including, for example, the Journal of the American Statistical Association, The Annals of Applied Statistics, and The American Statistician. Miguel’ research and trajectory has been recognized with a variety of awards—including the Lindley Prize from the International Society of Bayesian Analysis (ISBA). Miguel de Carvalho is an Elected Fellow of the International Statistical Institute (ISI), he is an Elected Member of the prestigious Council of the ISI, and a former member of the Board of Directors of ISBA.
Abstract: Extreme events, such as hurricanes of unprecedented strength, heatwaves surpassing historical temperatures, and floods inundating regions previously deemed safe, have become alarmingly frequent in recent years.
In this talk, I will highlight how Statistics and Data Science contribute to assessing the risk of extreme events, gauging their likelihood, and mitigating their impact on modern society. I will introduce methods from Extreme Value Theory and illustrate their applications using real data analyses. The methods that we will explore are being shaped by a community dedicated to extrapolating beyond observed data—into the tails of a distribution—drawing insights about the risk of extreme events, and understanding the dynamics governing extreme values across time and space.
This talk is based on:
– Coles, S., de Carvalho & Davison, A. C. (in preparation) An Introduction to Statistical Modeling of Extreme Values. Second edition. Springer: New York.