Matthew Nunes

This fully-funded PhD position at the University of Bath offers an exciting opportunity to develop new statistical tools for network-structured time series data. Large, multivariate time series are increasingly prevalent across diverse fields such as biology, medicine, social media, cyber security, and finance. In many cases, these data are observed on the nodes or edges of a network, or a network structure can be inferred to enhance model sparsity and inference. The project aims to create mathematically rigorous analysis tools tailored to capture complex data dynamics in various scenarios, addressing open methodological and computational challenges in statistical analysis. The successful candidate will contribute to the development of statistical models for evolving, interconnected stochastic processes, leveraging information at network nodes and edges. The modelling framework will focus on identifying key relationships among large sets of time series, with applications in forecasting, anomaly and extreme event detection, and classification of node or edge states. These innovative tools will be applied to generate impactful insights in both scientific and industrial contexts. This PhD is aligned with the EPSRC-funded Network Stochastic Processes and Time Series (NeST) research programme, which unites researchers from the Universities of Bath, Edinburgh, Oxford, York, Imperial College London, and the London School of Economics and Political Science, alongside industrial and government partners such as BT, EDF, and the Office for National Statistics. The student will join Professor Matthew Nunes at the University of Bath node, become part of the dynamic NeST team, and engage in cross-institutional collaboration. Applicants should be self-motivated and passionate about developing methodological and computational statistical techniques to solve practical problems. Strong mathematical, statistical, and programming skills are essential. Candidates must hold, or expect to receive, a First Class or good Upper Second Class UK Honours degree (or equivalent) in a relevant subject. A master’s qualification is advantageous. Non-UK applicants must meet the English language requirement by the application deadline. Funding is provided through a University of Bath studentship, tenable for 3.5 years, covering tuition fees, a stipend (£21,805 per annum in 2026/7), and access to a training support budget. The position is based in the Department of Mathematical Sciences at the University of Bath. To apply, submit a formal application via the University of Bath’s online application form for a PhD in Statistics before the deadline. In the 'Funding your studies' section, select 'University of Bath URSA' as the studentship. In the 'Your PhD project' section, quote the project title and the lead supervisor's name. Informal enquiries are encouraged and should be directed to Professor Matthew Nunes ([email protected]). Applications may close earlier than the advertised deadline if a suitable candidate is found, so early submission is recommended. The University of Bath values diversity and inclusion, welcoming applications from under-represented groups. If you have circumstances affecting your educational attainment, you are encouraged to mention them in your application. Reference: Knight, M. I, Leeming, K., Nason, G. P. and Nunes, M. A. (2020) Generalised Network AutoRegressive Processes and the GNAR package. Journal of Statistical Software, 96 (5), 1-36.

This PhD project at the University of Bath, in collaboration with Renishaw Plc, focuses on developing spatial artificial intelligence methods for hyperspectral image analysis. Hyperspectral imaging produces high-resolution images across hundreds of spectral wavelengths, offering significant potential for scientific and industrial applications such as medical diagnosis and geophysical mineral exploration. However, current machine learning approaches often lack spatial awareness and are computationally intensive, limiting their practical use. The research aims to create efficient, spatially-aware machine learning and mathematical optimisation techniques to enhance hyperspectral image analysis. Key research questions include leveraging spatial information for improved analysis, exploiting spectral correlations, overcoming computational bottlenecks, and quantifying uncertainty to support statistical decision-making. The project will involve aspects of machine learning, statistics, signal processing, and data dimension reduction, with the goal of providing practical guidelines for practitioners. The successful candidate will have the opportunity to undertake a placement at Renishaw Plc, gaining valuable industry experience and access to technical expertise and data. The Department of Mathematical Sciences at the University of Bath offers a supportive and inclusive research environment, welcoming applicants from diverse backgrounds. Funding: The position is funded for 4 years, covering tuition fees, an enhanced stipend (£23,805 per annum for 2026/7), and a training support budget. The studentship is open to both Home and International students, though international applicants must cover their own relocation, visa, and healthcare surcharge costs. Eligibility: Applicants should hold, or expect to receive, a First Class or good Upper Second Class UK Honours degree (or equivalent) in a relevant subject. A master’s degree is advantageous. Non-UK applicants must meet the English language requirement by the application deadline. Application Process: Apply online for a PhD in Mathematical Sciences at the University of Bath. Select ‘EPSRC-DTG’ as the studentship and quote the project title and lead supervisor’s name. Early application is recommended as the position may close before the advertised deadline if filled. Contact Professor Matthew Nunes for informal enquiries. Deadline: 30 April 2026. Note that the application system will be unavailable from 11–15 April 2026.