I am a Principal Lecturer in Statistics at the University of Brighton, where I serve as Course Leader for the BSc in Data Science and the MSc in Artificial Intelligence and Machine Learning. My research specialises in generalised estimating equations (GEE) and statistical methods for longitudinal and correlated data and high-dimensional analysis. My work spans methodological research and open-source software development, with several R packages published on CRAN and cited in textbooks.

I trained in Statistics at the University of Florida (PhD, under Alan Agresti) before joining the University of Cambridge and EMBL-EBI as a joint postdoctoral fellow, working on high-dimensional and genomic data with Simon Tavaré and John Marioni. I have applied these methods to real-world datasets across health and clinical research settings, and hold NIHR Good Clinical Practice (ICH E6(R3)) certification.

My research centres on statistical methods for correlated and longitudinal data, particularly generalised estimating equations (GEE) for multinomial, binary, and count responses, including bias-reduction and penalisation techniques to address separation and small-sample bias. I also work on high-dimensional statistics, including covariance matrix estimation and hypothesis testing for multivariate and genomic data.

My research is motivated by cross-disciplinary collaborations with clinicians, biologists, and bioinformaticians, and has been applied to health studies, clinical datasets, and omics data, alongside applications in the social sciences. I disseminate the methods I develop as open-source R packages on CRAN and Bioconductor, with an emphasis on numerical accuracy and rigorous software validation, the same standards expected of statistical software in regulated research environments.

I teach a wide range of undergraduate and postgraduate statistical modules at the University of Brighton, to both specialist and non-specialist audiences. My approach prioritises genuine understanding of statistical concepts over memorisation of technical procedures, encouraging independent, critical thinking rather than rote application of methods.

A recurring theme in my teaching is translating statistical ideas for audiences without a mathematics background, a skill that carries directly into cross-functional collaboration outside academia. I incorporate contemporary statistical software, including R and SAS, into my teaching material so students build practical, workplace-relevant technical skills alongside theoretical understanding.

I am interested in supervising postgraduate research students in machine learning methods, categorical data analysis, multivariate analysis, and high-dimensional statistics, with a particular interest in methods for correlated and longitudinal data. Projects typically involve developing open-source software — often destined for CRAN or Bioconductor — to implement, validate, and disseminate research findings, giving students direct experience in reproducible, production-quality statistical software development alongside the underlying methodology.

I actively translate statistical methodology into practical applications beyond academic publication. As a Co-Investigator on the DRIVA project (Digital Research & Innovation Value Accelerator, funded by the European Regional Development Fund), I developed gatwickAPI, an R package enabling small and medium-sized enterprises to access and analyse Gatwick Airport's real-time flight data, and delivered training sessions on user experience design for participating businesses.

I have led a statistics consultancy team providing methodological support to the University of Brighton's Doctoral College and to Brighton and Sussex Medical School, working directly with researchers and clinicians to design and analyse studies. I have also served as External Examiner for the MSc in Applied Statistics at Birkbeck, University of London, contributing to quality assurance in statistics education more broadly.