About the project
Climate and political shocks can trigger clustered defaults across interconnected financial networks. This PhD project will develop deep learning models and network-based stress testing tools to assess how such events impact credit risk and financial stability, using real nationwide microentrepreneur data in a partnership with a leading microfinance institution.
Climate and political events are increasing in frequency and intensity, posing significant challenges for financial stability. When firms are regionally concentrated or interconnected through supply chains, shocks can propagate across the financial system and lead to clusters of defaults. Existing stress-testing approaches typically assess risk at the portfolio level and overlook these network effects, leaving a major methodological gap—and a clear research opportunity. This PhD project aims to advance the modelling of systemic risk in complex financial networks. First, you will design deep learning architectures capable of capturing dynamic borrower networks and predicting how climate and political shocks influence credit risk and default propagation. Second, you will develop a network-based climate and political stress-testing framework, offering financial institutions a more realistic tool for evaluating system-wide vulnerability. The project will be grounded in real-world data through a collaboration with a leading microentrepreneur lender globally. You will work with regional and supply-chain network data, loan-level information, and high-resolution climate and weather indicators to evaluate the proposed models and stress-testing strategies in nationwide scenarios. This research has the potential to influence risk management practices in financial institutions, inform regulatory approaches to systemic climate and political risk, and contribute cutting-edge methods to the academic fields of financial networks, machine learning, and climate finance. You will join an active research environment, engaging with experts in network science, AI, and financial risk modelling.
Additional Information:
You will also be supervised by organisations other than the University of Southampton, including Prof. Cristián Bravo from Wester University (Canada).
