
Post-Graduate Stories
Jake Spertus Will Work on Methodologies for “Detection of Offline and Online Change-points with Applications in Public Health and Finance”
As a 2025 CANSSI-StatLab Distinguished Postdoctoral Fellow, Jake Spertus will take part in a comprehensive program that involves teaching, interdisciplinary or applied collaboration, professional development, and a research project focusing on the development of new change-point methodologies for multivariate time series, for either offline or online data, with applications in public health and actuarial science. Jake will work under the supervision of Professor Bouchra Nasri (Université de Montréal) and Professor Yunhong Lyu (Trent University).
Program: CANSSI-StatLab Distinguished Postdoctoral Fellowship
Region: National
Date: 2025–2027
Project Focus Areas
Jake Spertus will contribute to the development of new change-point methodologies for multivariate time series, for either offline or online data. These methodologies can be used to build early warning systems for infectious diseases, taking into account observations from one or more regions. For financial time series, they can be used to detect changes in the multivariate distribution or dependence that indicate when a proposed model is no longer valid, which is very important for investment strategies and applications in actuarial science. Jake will oversee the theoretical and computational results of the work and will be encouraged to create R packages implementing the proposed methodologies.
The project will provide Jake with access to public health data used to monitor and understand the behaviour of specific public health threats, as well as financial data available publicly or through collaborators at the National Bank of Canada.
In addition to conducting research, Jake will mentor undergraduate students and potentially graduate students to acquire supervisory experience.


Getting to Know Jacob
Jake Spertus is a Postdoctoral Researcher in the Department of Statistics at the University of California, Berkeley. He received his PhD in Statistics at Berkeley in May 2024 under the supervision of Professor Philip B. Stark. His doctoral research focused on developing methods for design-based inference, particularly finite-sample nonparametric tests for stratified or sequential designs.
Before Berkeley, Jake worked as a Research Assistant in the Department of Health Care Policy at Harvard Medical School in the lab of Sharon-Lise Normand. He received his BA in Mathematics at Bowdoin College.
He is enthusiastic about a wide range of problems with potentially positive impacts on science or society and has worked on applications in risk-limiting post-election audits; soil organic carbon measurement; cardiovascular and psychiatric outcomes research; graduate school admissions; and motor carrier safety, active transportation, and infrastructure planning.
His technical research interests include causal inference; nonparametrics; design-based statistics; policy optimization; sequential analysis; and supervised learning.
Jake sees the CANSSI-StatLab fellowship as a valuable step in his career development.
“I would like to continue to develop statistical methods, collaborate on scientific research, and teach data science. The CANSSI fellowship [will] expose me to new interdisciplinary problems and statistical methods, support collaborations with Canadian and international researchers, and provide opportunities for me to receive formal pedagogical training or gain hands-on experience teaching in new environments. Ultimately, I plan to apply for a professorship in the United States or Canada, and the fellowship [will] set me up well for this next step.”
The CANSSI fellowship [will] expose me to new interdisciplinary problems and statistical methods, support collaborations with Canadian and international researchers, and provide opportunities for me to receive formal pedagogical training or gain hands-on experience teaching in new environments.
About the Supervisors
Bouchra Nasri
Bouchra Nasri is a faculty member (Associate Professor) in Biostatistics in the Department of Social and Preventive Medicine at the Université de Montréal. Professor Nasri holds a Fonds de recherche du Québec – Santé (FRQS) Junior 2 award in Artificial Intelligence in Health and Climate Change and is a principal investigator on grants funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Canadian Institutes of Health Research (CIHR) in theoretical statistics for complex data and mathematical modelling for infectious diseases. She is a Co-lead of the data management theme (2021–2024) of the One Health Modelling Network for Emerging Infections (OMNI) and a member of Mathematics for Public Health (MfPh), funded by NSERC and the Public Health Agency of Canada (PHAC). Since March 2023, she has been nominated as Chair of PathCheck’s Data Informatics Center of Epidemiology, and since 2024, she has been a Co-director of the Digital Health Network in Québec. Professor Nasri has authored or co-authored papers on time series, dependence modelling, multivariate statistics, mathematical modelling for infectious diseases, text mining, and evidence synthesis. Her lab focuses on developing models for infectious diseases and public health threats.
Yunhong Lyu
Yunhong Lyu is an Assistant Professor in the Department of Mathematics at Trent University. She earned her PhD in Statistics with expertise in applied probability and stochastic processes in the Department of Social and Preventative Medicine at the Université de Montréal. Her research interests include statistical inference, statistical financial modelling, applied probability, stochastic processes, and data analysis.