Arnold FADIKPE

Designed and backtested systematic models to replicate book PnL using index options and Lasso regression, improving signal stability and predictive accuracy across market regimes. Engineered high-frequency datasets and APIs for real-time and historical multi-regional portfolio risk analysis integrating KDB+ and Polars.

Computed daily variance dispersion and implied correlation signals to identify systematic opportunities for volatility traders and onboarded analytics into research datasets. Quantified vol-surface stability with fit-to-market and arbitrage error measures and maintained real-time scenario and volatility-targeting trading GUIs.

Conducted event-driven statistical research on large-scale point-in-time datasets of US share buyback programs and identified alpha-generating patterns in post-announcement return profiles.

Developed deep-learning portfolio optimization models minimizing asset concentration while tracking CAC40 performance and integrated autoencoder architectures tuned with Optuna.

Completed an MSc in Applied Mathematics in Finance and Data Science with focus on quantitative methods and data science for finance.

Completed an integrated BSc-MSc program in Applied Mathematics in Finance emphasizing quantitative finance and mathematical modeling.

Completed MPSI followed by MP* preparatory classes with a focus on advanced mathematics and problem solving.