Bayesian & Computational Statistician | Multivariate, Survival & Missing-Data Methods | High-Dimensional Data
I am a Lecturer of Statistics at Columbus State Community College and a statistician developing Bayesian, computational, and multivariate methods for complex, heterogeneous, and high-dimensional data.
My research focuses on reliable statistical inference in the presence of censoring, missingness, dependence, heterogeneity, and uncertainty, with applications to biological and biomedical data.
- Bayesian & Computational Statistics: hierarchical modeling, adaptive prior learning, MCMC, NUTS, simulation-based inference, and uncertainty quantification
- Multivariate & High-Dimensional Methods: dependence modeling, correlated outcomes, heterogeneous data, and statistical learning
- Survival, Longitudinal & Missing-Data Methods: censored lifetime data, frailty and joint models, informative missingness, and repeated measurements
- Biological & Biomedical Data Science: high-dimensional proteomics, multi-omics data, survival outcomes, and computational statistical methodology
Computational materials for adaptive Bayesian Weibull failure-time modeling using R, Stan, MCMC, and the No-U-Turn Sampler (NUTS).
Associated publication:
Oketch, T., & Sepehrifar, M. (2026). Modeling complex life systems: Bayesian inference for Weibull failure times using adaptive MCMC. Statistical Papers, 67(1), Article 2.
Published Article · arXiv Preprint
Computational research on missing-value imputation for high-dimensional mass spectrometry-based proteomics, including Bayesian MCMC, MICE, quantile-regression-based methods, and machine-learning approaches.
Programming: R · Python · SAS · SQL
Bayesian Computing: Stan · MCMC · NUTS
Methods: Survival Analysis · Multivariate Analysis · Missing-Data Methods · High-Dimensional Inference
Research Computing: Simulation Studies · Reproducible Workflows · Statistical Visualization · Computational Model Evaluation
I am extending my work toward adaptive Bayesian survival models, multivariate dependence structures, informative censoring and missingness, longitudinal and joint modeling, and high-dimensional biological and biomedical applications.
I welcome methodological and interdisciplinary collaborations involving Bayesian computation, survival and longitudinal analysis, multivariate methods, missing-data methodology, and complex biological or biomedical data.