The Statistical Society of Canada (SSC) has awarded the 2026 Canadian Journal of Statistics Award to University of Waterloo researchers Meixi Chen, Reza Ramezan and Martin Lysy for their paper, Fast and scalable inference for spatial extreme value models, published in the Canadian Journal of Statistics in 2025. Lysy and Ramezan are faculty members in the Department of Statistics and Actuarial Science, while Chen was completing her PhD in Statistics at the time of writing the paper before joining Morgan Stanley as a Quantitative Strategist after graduation.
Presented annually, the award recognizes an outstanding article published in the journal during the previous year, honouring excellence in both methodological innovation and presentation.
The award-winning paper addresses a key challenge in modeling extreme weather events across large geographic regions. Understanding the likelihood and impact of rare events such as severe storms, heat waves, and flooding is increasingly important for climate risk assessment, reinsurance, power grid planning, and agricultural forecasting.
The researchers developed a computational framework that enables sophisticated spatial extreme value models to be applied to much larger datasets while maintaining a high level of accuracy. The approach dramatically reduces the computational resources required, making it practical for researchers to build, compare, and refine multiple models.
“Our goal was to make state-of-the-art models for weather extremes practical to use on large datasets,” said Dr. Lysy. “What previously could take more than a month of computation can now be completed in less than a day, making it feasible for researchers to compare and refine models in ways that were previously out of reach.”
The paper was recognized for its clear presentation and important contribution to spatial extreme value analysis, advancing researchers’ ability to study and better understand rare and extreme weather events.