Conference: EVA 2025: The 14th International Conference on Extreme Value Analysis

The project will provide financial support for junior researchers to attend the 14th International Conference on Extreme Value Analysis (EVA 2025) to be organized and hosted at the University of North Carolina at Chapel Hill from June 23 to June 27, 2025. Extreme value analysis refers to the statistical modeling of extreme data observations, allowing to make probabilistic predictions beyond the range of observed data.

With the mathematical foundation going back to the mid 20th century, the scope of EVA expanded considerably over the last decades, including applications in insurance, finance, engineering, atmospheric science, and other domains. The EVA conferences are bi-annual, and they attract some of the most prominent researchers in statistics and probability, as well as more applied domains working on extremes.

The supported junior researchers will have opportunities to attend a satellite workshop with tutorials, to present their work in oral or poster sessions, and to participate in a data challenge and best student paper competition.

EVA started with univariate models and frequentist methods for independent and identically distributed observations, but has since seen extensions to time series, spatial and spatial-temporal data, Bayesian methods, and so on. It has had a profound impact on a wide range of applications in atmospheric science (droughts, floods), finance (value-at-risk, market crashes), engineering (reliability, rare event prediction), social networks (viral tweets) and other domains.

Over the recent past, the field's impact has grown by making and exploiting connections to high-dimensional statistics (graphical models, dimension reduction), machine learning (prediction of extremes) and AI (training of neural networks) methods. These various themes, both classical and modern, will be well represented at the conference EVA 2025. The conference website can be found at https://eva2025.unc.edu.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.