The classical modeling of spatial extremes relies on asymptotic models (ie, max‐stable or r‐
Pareto processes) for block maxima or peaks over high thresholds, respectively. However, at …

Predictions and forecasts of machine learning models should take the form of probability
distributions, aiming to increase the quantity of information communicated to end users …

The health consequences of climate variability and change are diverse, potentially affecting
the burden of a wide range of health outcomes, including illnesses and deaths related to …

In multivariate or spatial extremes, inference for max-stable processes observed at a large
collection of points is a very challenging problem and current approaches typically rely on …

Predictions and forecasts of machine learning models should take the form of probability
distributions, aiming to increase the quantity of information communicated to end users …

We tackle the modeling of threshold exceedances in asymptotically independent stochastic
processes by constructions based on Laplace random fields. Defined as mixtures of …

To mitigate the risk posed by extreme rainfall events, we require statistical models that
reliably capture extremes in continuous space with dependence. However, assuming a …

Peaks-over-threshold analysis using the generalised Pareto distribution is widely applied in
modelling tails of univariate random variables, but much information may be lost when …

Extreme-value theory has been explored in considerable detail for univariate and low-
dimensional observations, but the field is still in an early stage regarding high-dimensional …

Likelihood-based procedures are a common way to estimate tail dependence parameters.
They are not applicable, however, in non-differentiable models such as those arising from …