AI-Enhanced Modelling of Macroeconomic and Financial Markets: Housing, Labour, and Financial Asset

With growing economic uncertainty worldwide, driven by geopolitical tensions, technological shifts, climate challenges, and volatile markets, there is increasing demand for more adaptive, data-driven approaches to understanding and predicting economic dynamics. At the same time, the rapid proliferation of Artificial Intelligence (AI) is reshaping how economists model uncertainty, analyse large-scale systems, and inform decision-making.

This PhD project investigates how modern AI and machine-learning methods can complement or challenge traditional economic approaches in environments characterised by nonlinear dynamics, regime shifts, and rapidly evolving signals.

The project provides substantial flexibility and can be tailored to the student’s interests within three broad application areas.

1. Housing markets and society

The housing sector is central to economic stability and social wellbeing. It also exhibits strong spatial heterogeneity, policy sensitivity, and cyclical behaviour, making it an ideal domain for methodological innovation.

Possible themes include:

• Price and rent dynamics, high-frequency prediction, and market cyclicality.
• Housing inequality, affordability, and the distributional consequences of market shocks.
• Social housing and public policy design, including allocation, sustainability, and long-term planning.
• Neighbourhood effects and community wellbeing, examining how the built environment influences social mobility, segregation, or labour outcomes.
• Environmental quality and living conditions, such as green space access, energy efficiency, and climate risks.

Students may draw on UK or international housing data, depending on the research focus.

2. Financial Markets and Asset Pricing

Modern financial systems generate vast volumes of data, making them ideal for testing new modelling approaches.

Potential research directions include:

• Asset price prediction and market regime classification using high-frequency or cross-market signals.
• Volatility modelling, tail risk analysis, and early-warning indicators.
• Capital flows and macro-financial linkages, especially during periods of instability.
• Policy interactions, such as how monetary and macro-prudential policies propagate through financial markets.

This strand is suitable for candidates interested in empirical finance, macro-finance, or risk analysis.

3. Methodological innovation: AI vs. traditional economic models

Across all topic areas, students are encouraged to explore how AI methods compare with or complement:

• Dynamic Stochastic General Equilibrium (DSGE) models.
• Overlapping Generations (OLG) models.
• Econometric approaches such as VARs, GMM, and state-space models.

Possible contributions include integrating deep learning into structural models, identifying nonlinear propagation mechanisms, or improving forecasting and counterfactual analysis.

4. Labour markets and economic behaviour

Labour markets face structural change due to automation, shifting demographics, and evolving work patterns.

Themes may include:

• Wage dynamics, labour mobility, and job-matching efficiencies.
• Predicting labour shortages or wage pressures using microdata or firm-level signals.
• Policy evaluation, including minimum wage reforms, training programmes, and migration policies.
• AI-enabled analysis of job postings, firm behaviour, and worker transitions.

This option suits students combining labour economics, applied microeconometrics, and machine-learning methods.

Data and approach

Depending on student preference, the project may involve:

• Empirical analysis using macro, sectoral, financial, or spatial datasets.
• Model development integrating AI techniques with economic theory.
• Policy-relevant analysis aimed at informing governments, firms, or NGOs.

Data can be drawn from international sources.