Recent advances in extreme value theory have established $$\ell $$-Pareto processes as the natural limits for extreme events defined in terms of exceedances of a risk functional. In this paper we provide methods for the practical modelling of data based on a tractable yet flexible dependence model. We introduce the class of elliptical $$\ell $$-Pareto processes, which arise as the limits of threshold exceedances of certain elliptical processes characterized by a correlation function and a shape parameter. An efficient inference method based on maximizing a full likelihood with partial censoring is developed. Novel procedures for exact conditional and unconditional simulation are proposed. These ideas are illustrated using precipitation extremes in Switzerland.
Spatial modeling of rare events has obvious applications in the environmental sciences and is crucial when assessing the effects of catastrophic events (such as heatwaves or widespread flooding) on food security and on the sustainability of societal infrastructure. Although classical geostatistics is largely based on Gaussian processes and distributions, these are not appropriate for extremes, for which max-stable and related processes provide more suitable models. This paper provides a brief overview of current work on the statistics of spatial extremes, with an emphasis on the consequences of the assumption of max-stability. Applications to winter minimum temperatures and daily rainfall are described.
We propose an approach to spatial modeling of extreme rainfall, based on max‐stable processes fitted using partial duration series and a censored threshold likelihood function. The resulting models are coherent with classical extreme‐value theory and allow the consistent treatment of spatial dependence of rainfall using ideas related to those of classical geostatistics. We illustrate the ideas through data from the Val Ferret watershed in the Swiss Alps, based on daily cumulative rainfall totals recorded at 24 stations for four summers, augmented by a longer series from nearby. We compare the fits of different statistical models appropriate for spatial extremes, select that best fitting our data, and compare return level estimates for the total daily rainfall over the stations. The method can be used in other situations to produce simulations needed for hydrological models, and in particular, for the generation of spatially heterogeneous extreme rainfall fields over catchments.
1 December 2016 in The Annals of Applied Statistics
The Brown–Resnick max-stable process has proven to be well suited for modeling extremes of complex environmental processes, but in many applications its likelihood function is intractable and inference must be based on a composite likelihood, thereby preventing the use of classical Bayesian techniques. In this paper we exploit a case in which the full likelihood of a Brown–Resnick process can be calculated, using componentwise maxima and their partitions in terms of individual events, and we propose two new approaches to inference. The first estimates the partitions using declustering, while the second uses random partitions in a Markov chain Monte Carlo algorithm. We use these approaches to construct a Bayesian hierarchical model for extreme low temperatures in northern Fennoscandia.