Tobias Oketch is a statistician, researcher, and educator whose work spans Bayesian, computational, and multivariate statistical methods for complex, heterogeneous, and high-dimensional data. His research interests include survival analysis, missing-data methodology, dependence modeling, statistical computing, and applications to biological and biomedical data.
He earned his Ph.D. in Mathematical Sciences with a concentration in Statistics and his M.S. in Statistics from Mississippi State University. His doctoral research focused on adaptive Bayesian methods and Markov chain Monte Carlo computation for survival and lifetime data, with particular attention to censored and heterogeneous samples. His research has also included computational methods for missing-value imputation in high-dimensional mass spectrometry-based proteomics data. These complementary areas have shaped his broader interest in developing statistical methods that account for heterogeneity, incomplete observations, complex dependence structures, and uncertainty.
Alongside his research, Tobias teaches statistics and mathematics at Columbus State Community College. His teaching emphasizes conceptual understanding, statistical reasoning, active learning, and the thoughtful use of computational tools and real-world applications to help students develop confidence as quantitative thinkers.
Ph.D., Mathematical Sciences (Statistics), Mississippi State University, August 2025
M.S., Statistics, Mississippi State University, August 2025
M.S., Mathematical Sciences, East Tennessee State University, May 2017
B.S., Applied Statistics, Maseno University, December 2008
Tobias's long-term research goal is to develop robust, interpretable, and computationally efficient statistical methodology for complex scientific data. He is particularly interested in integrating Bayesian computation, multivariate modeling, dependence structures, missing-data methodology, and high-dimensional inference to address problems involving correlated and heterogeneous outcomes.
His broader research vision includes applications in biological, biomedical, and environmental sciences, where modern studies increasingly involve multiple data types, incomplete observations, longitudinal measurements, and complex dependence. He welcomes interdisciplinary collaborations that connect methodological statistical research with substantive scientific questions.