Contributions of PRA to Probability and Statistics: A Methodological History
A Talk Dedicated to the Memory of Corwin C. Atwood (2026)

Probability Theory and Statistics are foundational disciplines within the risk sciences, alongside the broader field of Uncertainty Quantification. While Probabilistic Risk Assessment (PRA) has historically drawn on established methods from these areas, it has also generated significant original contributions that have advanced all three fields.
This talk presents a methodological history of these contributions, examining both their development and their broader impact on other disciplines, including Decision Theory, Data Science, and Artificial Intelligence. Key contributions include the formal distinction between aleatory and epistemic uncertainty; approaches to modeling both parameter and model uncertainty; developments in two-stage and empirical Bayesian estimation; structured methods for the elicitation and use of expert judgment; techniques for incorporating partially relevant data into parameter estimation; the data mapping method; and advances in uncertainty propagation and Monte Carlo sampling.
The presentation will also highlight ongoing challenges and identify emerging frontiers for future research.