About
I am a Research Associate in Statistics at the University of Glasgow. My current work focuses on modelling carbon emissions associated with UK household energy use, in partnership with NatWest Group.
I originally trained in physics, with a particular focus on gravitation and gravitational-wave data analysis. I received the Quantum Fields and Fundamental Forces MSc from Imperial College London in 2017, and completed my PhD in Gravitational Wave Data Analysis at the University of Glasgow in 2022. My PhD work combined Bayesian inference, gravitational-wave modelling and efficient likelihood computation, and was awarded the Kelvin Award.
Since joining the School of Mathematics and Statistics in 2024, my research has moved into applied statistical modelling for energy and environmental data. I work with household-level energy consumption, property and emissions data to understand how accurately domestic carbon emissions can be estimated under different levels of data availability.
Research interests
My current research focuses on statistical modelling of household energy use and carbon emissions, particularly in the context of residential mortgage portfolios and financed-emissions reporting.
In my current role as a Research Associate in Statistics at the University of Glasgow, I work with linked household energy and property data — including smart-meter gas and electricity consumption, Energy Performance Certificate data, dwelling characteristics and national electricity carbon-intensity data — to estimate annual household emissions. A central aim of this work is to understand how accurately emissions can be estimated under different levels of data availability, and what this implies for data-quality frameworks such as the PCAF data-quality score hierarchy.
Applied emissions modelling
Estimating household energy-related emissions from imperfect, heterogeneous property and consumption data.
Data quality and uncertainty
Testing how data availability, imputation and model choice affect out-of-sample predictive accuracy.
Methodologically, this work involves data curation, harmonisation and integration, regression modelling, additive models, imputation, model selection, cross-validation, and out-of-sample predictive-error analysis. I am particularly interested in the relationship between data quality, model accuracy and uncertainty: for example, whether higher-quality input data actually leads to meaningfully better emissions estimates in practice.
Although my current applied work focuses on household energy and emissions modelling, my broader statistical background is in Bayesian inference and gravitational-wave parameter estimation. My PhD work developed computationally efficient likelihood and downsampling methods for long-duration gravitational-wave signals, particularly in the context of the future LISA mission. This background continues to shape my interest in uncertainty quantification, likelihood-based inference and efficient statistical computation.
More broadly, I am interested in Lorentzian geometry, general relativity and variational principles, particularly in the context of Lorentzian manifolds. I also have interests in information geometry, gravitational-wave astronomy and data analysis, and the foundations of spacetime physics. Outside mathematics and statistics, I have a strong interest in music and composition, and hold a BMus degree in composition from the Royal Academy of Music.