MSc Mathematical Trading & Finance (Bayes Business School) · BSc Mathematics & Philosophy (Durham)
President, Quantitative Finance Society | Quant researcher focused on systematic strategies, risk modelling, and financial econometrics
I design and validate interpretable models under uncertainty, from structured product pricing and volatility forecasting to market regime detection.
My work emphasises economic intuition, reproducibility, and statistical robustness, bridging advanced mathematics with real-world financial applications.
- Regime-aware systematic strategies across equities, fixed income, and macro
- Quantitative model design, validation, and backtesting under structural constraints
- Cross-disciplinary methods for robustness and inference (stochastic processes, ML, signal processing)
| Project | Summary | Key Results |
|---|---|---|
| Market Regime Identification Pipeline | Novel pipeline using Hilbert–Huang transforms, fractional differencing, and Wasserstein clustering to uncover economic/structural market regimes. | Identified 4 distinct regimes across global indices; robustness validated (median ARI = 0.67, perfect = 1 in conditions); outperformed HMMs and vol-based rules in crisis capture. |
| Deutsche Bank Quant Challenge (repo forthcoming) | Co-founded & led Bayes x DB challenge to replicate equity market-neutral strategies under real-world constraints; presented to Global Head of Quant Research. | Full pipeline delivery & presentation; emphasised collaboration, reproducibility, and practical risk controls. |
| Structured Bond Valuation & Hedging | Priced BNP Paribas capped/floored FRN in QuantLib; calibrated with cap vols & CDS spreads. | Found 1.52% undervaluation after CVA; optimised hedging reduced DV01 exposure by 92%. |
| Backtesting VaR Models | PIT & coverage tests across six VaR models. | Calibrated well at 90% but broke at 99%; Gaussian underpredicted tail risk (78 vs 37 breaches). |
| Adjusted Momentum Strategy | Replicated Lou & Polk’s comomentum model via Fama–MacBeth regressions. | Sharpe ratio rose 0.75 → 0.78, but not statistically significant. |
| Portfolio Forecasting & Optimisation | EGARCH(1,1)-guided allocation model and mean-variance portfolios. | Forecasting model achieved 4.12% CAGR, +1.8% over FTSE benchmark in backtest. |
- Languages/Libraries: Python (scikit-learn, TensorFlow, QuantLib), MATLAB, LaTeX
- Financial Modelling: Risk analysis, option pricing, portfolio optimisation, econometrics
- Software/Infra: Git, Bloomberg Terminal, Jupyter, unit testing & reproducibility practices