معرفی
Joel Dubin is a Professor at the University of Waterloo with a joint appointment in the Department of Statistics and Actuarial Science and the School of Public Health Sciences. He leads the Health Data Science Lab (HDSL), focusing on high-dimensional health data analysis and predictive modeling for critical care, electronic health records, and population health.
- Education: MSc in Applied Statistics (Villanova University, 1993), PhD in Statistics (UC Davis, 2000)
- Previous Affiliations: Yale University (Assistant Professor), University of Texas M.D. Anderson Cancer Center, Veteran Affairs Health Services
His methodological research encompasses multivariate longitudinal data correlation, change point models, latent response models, and similarity-based prediction. Application domains span intensive care, aging studies, mobile health, cancer, nutrition, and environmental health.
Recent publications (2024-2025) highlight his work on causal inference for missing exposures, network treatment effects, dyadic network modeling, and personalized predictive models in contexts like COPD-19 pandemic analysis, mental health, and adolescent behavioral studies. Methodological innovations include noisy matrix completion and dynamic treatment regimes.
Labs & Collaborations: HDSL members specialize in ICU data analytics, causal inference, functional data analysis, and machine learning for health outcomes. The lab collaborates across disciplines in public health, medicine, and environmental sciences.
