Dr. Paul Marriott

Biography

Professor Paul Marriott joined the Department of Statistics and Actuarial Science at the University of Waterloo in 2004. After graduating in 1989, he held positions in statistics departments across the U.K., Singapore, and the U.S., including Surrey University, the National University of Singapore, and Duke University. His research interests span both theoretical and applied aspects of mathematical statistics, particularly in applying geometric ideas—often involving differential, affine, and convex algebraic geometry—to statistical theory. His early work focused on preferred point geometry, which found practical applications in econometrics. Notably, he introduced the concept of local mixtures, which exhibit desirable statistical properties and have applications in measurement error modeling, analysis of censored data, influence analysis, and dispersion modeling. Recently, he has concentrated on developing tools within the framework of information geometry to enhance the computational aspects of these theories, directly applying them to modeling. His applied research covers a diverse range of projects in fields such as physics, linguistics, medical statistics, and neuroscience, with recent studies analyzing neural spike train and EEG data.

Research Interests