Overview
I am a PhD student in Statistics at Texas A&M specializing in hierarchical modeling, uncertainty quantification, and model validation. I value collaborative problem-solving, mentorship, and communicating statistical work with clarity and transparency.
Check out our recent publication: Bayesian disease modeling with latent class analysis
Experience
- Develop Bayesian hierarchical and latent-variable models for prevalence estimation and diagnostic testing without a gold standard.
- Design simulation studies to evaluate identifiability, parameter recovery, robustness, and uncertainty quantification.
- Develop a latent severity extension to reduce model complexity and improve estimation in sparse-data settings.
- Serve as instructor for an undergraduate introductory statistics course covering probability, statistical inference, hypothesis testing, and regression.
- Teach students to perform exploratory and inferential statistical analyses using R and interpret results in real-world contexts.
- Develop and deliver course materials, assessments, and statistical computing instruction for an asynchronous online course.
- Developed and deployed credible intervals for predictive models, improving reliability of forecast.
- Designed repeatability and reproducibility experiments, strengthening product validation.
- Built predictive models linking chemical composition to transmission fluid performance to guide the data-driven product development decisions.
- Applied high-dimensional regression and clustering methods to DNA methylation data, identifying biomarkers linked to disease pathways.
- Analyzed NHANES data to evaluate dietary risk factors for hypertension in pregnancy, contributing to peer-reviewed publications.
Projects
Selected work in experimentation, causal inference, Bayesian modeling, and applied statistical analysis.
A/B Testing and Experimental Design
- Analysis of randomized product experiments, including metric definition, power and sample size calculations, treatment-effect estimation, uncertainty quantification, and experiment diagnostics.
Double/Debiased Machine Learning for Causal Inference
- Applications of Double Machine Learning for causal effect estimation, including randomized trials with non-compliance and observational data with high-dimensional confounding.
Bayesian Latent Severity Modeling
- Bayesian latent class modeling extended with a latent severity variable to reduce model complexity and improve estimation of disease prevalence and diagnostic test accuracy when data are limited and no gold standard test is available.
Geospatial Clustering
- Geospatial analysis of New York City taxi trips to identify spatial and weekly patterns in passenger demand.
Zero-Inflated Negative Binomial Regression
- Modeling overdispersed and zero-inflated infection counts to identify important predictors and characterize variation in count outcomes.
Publications
Nature Communications, 16:11600 (2025)
BMC Biology, 21:30 (2023)
Nutrients, 14(12):2545 (2022)