My research develops statistical methodology and applies it across disciplines, reflecting an interest in how general statistical thinking can serve as a foundation for enquiry in many different fields. My work covers parametric and non-parametric estimation, including distribution theory and the estimation of heavy-tailed, Pareto-type distributions, which underpins my ptsuite package for Pareto tail estimation. It also extends to computational statistics, and simulation studies. My doctoral research was in statistical genetics, where I developed novel analyses to reveal latent genetic architecture and improve genomic prediction accuracy for complex traits and diseases, building polygenic risk score models, designing methods for uncertainty estimation and confidence intervals, and developing models for gene and environment interaction analysis. Across these areas, my interest is in building methods that are both statistically sound and useful to researchers working on real problems.

A central part of my work is the development of open-source statistical software that makes these methods usable by others. I am the main author of three R packages: GPflexViz (2025) for visualising genomic analytics, iPRSue (2025) for polygenic risk score analysis with uncertainty quantification, and GxEprs (2023) for gene and environment interaction analysis, and a co-author of ptsuite (2019) for Pareto tail estimation. Alongside package development, my published work appears in journals across human genetics, genetic epidemiology, and applied statistics, reflecting collaborations that pair statistical expertise with domain knowledge from other fields. My statistical consulting experience reinforces this collaborative, applied orientation and continually surfaces new methodological questions.