My research focuses on developing multivariate, Bayesian, and computational statistical methods for complex, heterogeneous, and high-dimensional data. I am particularly interested in statistical frameworks that address dependence, missingness, censoring, and uncertainty while remaining computationally practical for modern scientific applications.
My work spans Bayesian inference, survival and longitudinal analysis, multivariate dependence modeling, missing-data methodology, statistical computing, and high-dimensional biological data analysis. Through this research, I aim to develop rigorous and interpretable statistical methods that improve inference and support scientific discovery in biological, biomedical, and other data-intensive fields.
Development of Bayesian methodology for hierarchical modeling, adaptive prior specification, posterior computation, uncertainty quantification, and simulation-based statistical inference.
Development of statistical frameworks for correlated outcomes and complex dependence structures, with particular interest in heterogeneous, incomplete, longitudinal, and mixed data.
Methodology for time-to-event, repeated-measures, censored, and incomplete data, including frailty structures, joint models, and informative missingness.
Statistical and computational methods for proteomics, multi-omics, correlated phenotypes, and other complex biological and biomedical data.
My research program integrates Bayesian and computational inference, high-dimensional biological data analysis, and multivariate modeling of complex and heterogeneous data.
My doctoral research focused on adaptive Bayesian methods for survival and lifetime data, particularly in small, heterogeneous, and censored samples. I developed Bayesian Weibull models combining adaptive prior specification with modern Markov chain Monte Carlo methods, including the No-U-Turn Sampler, to improve posterior inference and uncertainty quantification.
My continuing work investigates hierarchical survival models, frailty structures, informative censoring, adaptive prior learning, and joint modeling.
Through research at the Institute for Genomics, Biocomputing, and Biotechnology at Mississippi State University, I investigated missing-value imputation in mass spectrometry-based proteomics using MCMC, Multiple Imputation by Chained Equations, quantile-regression-based methods, and machine-learning approaches.
This work strengthened my interest in how missing-data methods affect not only predictive accuracy but also the preservation of variability and dependence structures important for downstream statistical inference.
Building on this foundation, I am expanding my research toward multivariate methods for correlated, heterogeneous, and incomplete outcomes. I am particularly interested in flexible dependence models that accommodate mixed data types, longitudinal observations, missingness, and high dimensionality.
Future directions include dependence modeling for correlated phenotypes, integrative analysis of biological and environmental data, and computational frameworks that combine survival, longitudinal, genomic, proteomic, and other complex outcomes.
A Hierarchical Bayesian Weibull Model under MNAR Censoring with Adaptive Prior Learning for Reliable Survival Inference in Small and Heterogeneous Samples
Development of a hierarchical Bayesian Weibull framework for survival inference under informative censoring, with adaptive prior learning designed for small and heterogeneous samples.
Adaptive Prior Learning in Bayesian Frailty and Joint Survival Models with Covariates: Theoretical and Inferential Implications
Investigation of adaptive prior specification in Bayesian frailty and joint survival models, with emphasis on covariate information, posterior uncertainty, and inferential performance.
Adaptive Bayesian Weibull Survival Modeling for High-Dimensional Multi-Omics Data with Structured Shrinkage and Biological Priors
Development of Bayesian Weibull survival models for high-dimensional multi-omics data using structured shrinkage and biologically informed prior specifications.
Performance Analysis of Computational Statistics Methods for Missing Value Imputation in Mass Spectrometry-Based Label-Free Quantitative Proteomics
Comparative evaluation of statistical and computational missing-data methods for high-dimensional proteomics, with emphasis on predictive accuracy, variance preservation, and downstream statistical inference.
I welcome methodological and interdisciplinary collaborations that connect statistical innovation with substantive scientific questions. I am particularly interested in collaborative projects involving biological, biomedical, environmental, and other data-intensive applications, as well as opportunities for methodological development, computational implementation, simulation studies, and reproducible statistical research.
I welcome collaborations with faculty, researchers, and students whose work would benefit from rigorous statistical methodology and computational approaches.
My research and teaching are mutually reinforcing. I incorporate statistical computing, simulation, real-world data, and contemporary methodological applications into my teaching, while teaching continually strengthens my commitment to clear statistical communication, reproducible analysis, and sound statistical reasoning.
Additional information about my publications, citations, research activities, and professional experience is available through: