School of Mathematics · University of Edinburgh
Ioannis Papastathopoulos
Lecturer in Statistics
My research centres on extreme value theory, developing probabilistic theory and statistical methods for rare events and their dependence across variables and time. Particular interests include geometric extremes, graphical models and conditional extremes, with applications to environmental risk and climate.
I have also worked at the interface of statistics and neuroscience, developing Bayesian point-process models for neural spike trains to understand how grid-cell firing encodes spatial and directional information.
Selected research
Graphical structure in geometric extremes
Understanding how extremal dependence propagates through a graph, with connections to conditional extreme value theory.
Papastathopoulos & Wadsworth · 2026 preprint
Geometric extremes and return sets
Radial generalised Pareto models and Bayesian inference for the probabilities of rare multivariate events.
Papastathopoulos, de Monte, Campbell & Rue · 2023 preprint, revised 2025
-
Hidden tail chains that describe how higher-order Markov processes evolve around an extreme event.
Papastathopoulos, Casey & Tawn · Advances in Applied Probability, 2025
Prospective PhD students
I welcome focused PhD enquiries from candidates with a strong background in probability, statistics and mathematics. Please briefly outline your research interests and how they connect with my work; prior experience in extreme value theory is not essential.
Academic roles and community
I am Director of Statistics Postgraduate programmes and a staff representative in the School of Mathematics. I am also a founding member of GL(EN)², the Glasgow–Lancaster–Edinburgh–Newcastle Extremes Network.
I was lead organiser of the 12th International Conference in Extreme Value Analysis (EVA 2021).
Contact
Room 4602, James Clerk Maxwell Building
School of Mathematics, University of Edinburgh
Peter Guthrie Tait Road, Edinburgh EH9 3FD, UK