A0394
Title: Estimation of spatiotemporal extremes via generative neural networks
Authors: Christopher Buelte - Ludwig-Maximilians-Universität München (Germany) [presenting]
Lisa Leimenstoll - Karlsruhe Institute of Technology (Germany)
Melanie Schienle - Karlsruhe Institute of Technology (Germany)
Abstract: Recent methods in modeling spatial extreme events have focused on utilizing parametric max-stable processes and their underlying dependence structure. A unified approach is provided for analyzing spatial extremes with little available data by estimating the distribution of model parameters or the spatial dependence directly. By employing recent developments in generative neural networks, a full sample-based distribution is predicted, allowing for direct assessment of uncertainty regarding model parameters or other parameter-dependent functionals. The method is validated by fitting several simulated max-stable processes, showing a high accuracy of the approach regarding parameter estimation, as well as uncertainty quantification. Additional robustness checks highlight the generalization and extrapolation capabilities of the model, while an application to precipitation extremes across Western Germany demonstrates the usability of the approach in real-world scenarios.
