How can early career faculty members in statistical sciences gain supervisory experience?
One option is CANSSI’s Collaborative Supervision Program (previously known as the Graduate Student Enrichment Scholarship), which gives early career faculty members an opportunity to co-mentor graduate students in statistical sciences alongside a more experienced faculty member. Program participants also receive up to $15,000 over two years to support enriched training experiences for the supervised student.
In 2026, the program will support five faculty members at universities stretching from Nova Scotia to British Columbia. We would like to congratulate these individuals and introduce them to the wider statistical sciences community:
Sumeet Kalia is an Assistant Professor in the Department of Statistics at the University of Manitoba whose research involves the development of novel biostatistical methods for epidemiology and health services research. He will work alongside Mohammad Jafari Jozani, a Professor in the same department, to co-supervise a PhD student (Ali Karoobi) investigating “High-dimensional Causal Inference with Case-base Sampling Using Machine Learning Methods.” The project will seek to answer two questions: 1. To what extent can case-base sampling improve the efficiency of causal effect estimation in continuous-time settings with high-dimensional covariates? 2. Does incorporating non-parametric machine learning methods (e.g., support vector machines) enhance causal effect estimation—measured by finite-sample bias and variance—relative to benchmark models such as the Cox proportional hazards model?
Théo Michelot, an Assistant Professor in the Department of Mathematics and Statistics at Dalhousie University with an interest in ecological statistics, statistical models of animal movement, stochastic processes, and time series analysis, will co-supervise a PhD student (Joseph Barss) working on “Multivariate Spatiotemporal Modelling of Interacting Animal Populations” alongside Joanna Mills Flemming, a Professor in the same department. This project will build on hierarchical (mixed effect) methodology to develop joint stock assessment models for multiple interacting species, using trawl surveys run by Fisheries and Oceans Canada (DFO) that yield large data sets of counts or biomass of different species over space and time.
Dingke Tang is an Assistant Professor in the Department of Mathematics and Statistics at the University of Ottawa with research interests that include causal inference, robust statistics, and large language models. He will co-supervise a PhD student (to be named) alongside David Haziza, a Professor in the same department. Their student will work on a project titled “Enhanced Causal Inference Methods Leveraging Rich Survey Sampling Data” to develop a unified framework that blends survey sampling theory with modern causal inference techniques to improve the reliability and interpretability of treatment effect estimation using population-based data sources.
Qinglong Tian is an Assistant Professor in the Department of Statistics and Actuarial Science at the University of Waterloo. His current research focuses on transfer learning, particularly addressing challenges related to distributional shift, out-of-distribution detection, and label noise. He approaches these problems through the lens of mixture models and density ratios. He will co-mentor a PhD student (Edward Chang) alongside Pengfei Li, a Professor in the same department. Their student will work on “ROC Curve Analysis and Optimal Biomarker Combination under Positive-Unlabeled Setting.” This project will address the challenge that arises in the use of biomarkers to identify diseased subjects in contaminated case-control studies when the case group contains diseased individuals, but the control group may also contain undiagnosed individuals.
Owen Ward is an Assistant Professor in the Department of Statistics and Actuarial Science at Simon Fraser University. His areas of interest are statistical and machine learning models for network data, statistical computation, and Bayesian statistics. His co-supervisor will be Joan Hu, a Professor in the same department. Their PhD student (Albert Shen) will address “Modelling Real World Dynamics with Temporal Point Process Models” with the goal of developing novel statistical methodology to infer the complex latent structure present in vast recurrent event data such as electronic communication, electronic health records, and financial transactions, along with computational tools and theoretical results for these methods.
Applications to CANSSI’s Collaborative Supervision Program are accepted from September 15 to November 30 each year.
