Olobatuyi, Kehinde

Olobatuyi, Kehinde
Assistant Professor

Mathematics ＆ Computer Science

Phone
(403) 329-2086
Email
kehinde.olobatuyi@uleth.ca

About Me

I am a Computational Statistician specializing in Bayesian statistics, machine learning, and public health data science. My work focuses on developing advanced statistical and computational methods, such as hybrid MCMC and high-dimensional optimization, to solve complex, real-world problems in infectious disease modelling and health analytics.

With a strong foundation in both theory and application, I am particularly interested in bridging methodological innovation with impactful public health outcomes. 

My research has addressed critical challenges such as COVID-19 under-reporting, contributing to evidence-based decision-making and policy. In addition to my research, I am deeply committed to teaching and mentorship. I support students and trainees in building strong analytical skills, emphasizing reproducible research, hands-on data analysis, and the integration of statistical theory with practical applications in Python and R.

I actively collaborate across disciplines and institutions, and I am passionate about developing scalable, data-driven solutions that advance healthcare systems.

My long-term goal is to contribute to a research-intensive academic environment while continuing to drive innovation at the intersection of statistics, machine learning, and public health.


Current Research

Bayesian Computation & Scalable Inference: Fast and reliable Bayesian inference for complex models, with emphasis on uncertainty quantification and computational efficiency.
State-Space, HMM & Latent-Process Models: Developing statistical models for temporal, longitudinal and multistate systems in which the underlying process is only partially observed.
Machine Learning & Deep Learning: Developing prediction and representation-learning methods that complement statistical modelling while preserving interpretability, calibration and uncertainty.

Infectious Disease, Health Systems & Health Data Science: Developing methods for learning from incomplete and heterogeneous health data to support surveillance, service planning and equitable decision-making.
Capture-Recapture, Population Enumeration & Hidden Populations: Developing principled methods to estimate populations and latent outcomes when data are incomplete, delayed or fragmented across administrative systems.

Publications

  1. K. Olobatuyi, J. Ma, P. Brown, and L. L. Cowen, “Multi-event dynamic capture-recapture model for big data: Estimating undetected COVID-19 cases in British Columbia, Canada,” Infectious Disease Modelling, vol. 11,
    pp. 764–786, 2026.
  2. O. Ariyo, K. Olobatuyi, and T. Baghfalaki, “A bayesian joint bent-cable model for longitudinal measurements and survival time with heterogeneous random-effects distributions,” Journal of Biopharmaceutical Statistics, pp. 1–14, 2025.
  3. K. Olobatuyi, S. Johns, M. Parker, H. S., and L. Cowen, “extbatchmarking: An R package for hidden Markov models for extended batch data.,” Accepted in Canadian Journal of Statistics, 2026.
  4. O. Ariyo, K. Olobatuyi, and I. Fwamba, “A multilevel joint model for hierarchical longitudinal and time-to-event data using an auxiliary mixed effects Poisson approach: Application to the scleroderma lung study,” Submitted to Journal of Lifetime Data Analysis, pp. 1–23, 2026
  5. Olobatuyi, K; Parker, M; Ariyo, O. (2023). Cluster-Weighted Model Based on TSNE algorithm for High Dimensional Data. International Journal of Data Science and Analytics. 17: 261–273. Published Refereed?: Yes, Open Access?: Yes

Degrees

  • 1. Postdoc funded by CIHR REDI Phase 1: 2023-2026
  • 2. Postdoc funded by UVIc Aspiration2023: 2022-2023
  • 3. Doctoral: University of Milano-Bicocca, Milan, Italy. 2017-2021
  • 4. Masters: Federal University of Agriculture Abeokuta. 2014-2016
  • 5. Bachelor: Federal University of Agriculture Abeokuta. 2008-2012

Personal Website

Curriculum Vitae