Research overview
I develop and apply Bayesian spatial modeling for public health, including environmental health and epidemiology. My work aims to bridge the gap between statistical methodology and applied sciences, and is motivated by real-world applications. As a statistician, I enjoy working in collaborative interdisciplinary research.
Research Interests:
- Spatial statistics
- Bayesian inference
- Causal inference
- Statistics education
Application Areas:
- Environmental Health
- Spatial Epidemiology
- Public Health & Health Services
Current projects
Bayesian spatial source apportionment for PFAS contamination
The goal of this project is to analyze groundwater wells from the GAMA dataset in California to attribute PFAS contamination to specific industrial sources using Bayesian hierarchical modeling. To address heavily skewed PFAS concentrations, we leverage the extended generalized Pareto distribution (EGPD) to accurately model both the lower and upper tails.
Upcoming: To be presented at the 2026 Joint Statistical Meetings, Boston, MA in August 2026.
Collaborators: Sean O’Connor, Ana Rappold, Brian Reich
Methods: Bayesian hierarchical modeling, spatial statistics, extreme values
Status: Ongoing
Modeling Cancer Care Access and Liver Cancer Mortality
This project implements a Bayesian spatial framework to evaluate how geographic isolation from specialty care impacts liver cancer outcomes.
Collaborators: Michela Fabricius, Sam Berchuck, Lisa McElroy, Norine Chan, Brian Reich, Allison Martin
Methods: Spatial statistics, Bayesian modeling
Status: Ongoing
Spatial Confounding for Discrete and Areal Data
This project aims to address spatial confounding for both discrete (binary and count) outcomes and areal spatial data.
Upcoming: To be presented by Nate Wiecha at the 2026 Joint Statistical Meetings, Boston, MA in August 2026.
Collaborators: Nate Wiecha, Sam Berchuck, Youngsoo Baek, Brian Reich
Methods: Spatial statistics, generalized linear models, causal inference
Status: Ongoing