Distributional crypto-return forecasting via Wasserstein-geodesic extrapolation in quantile-function space. WGeo family wins 12/12 (asset × horizon) cells over 6.75y walk-forward CRPS vs GARCH and classical baselines. v0.4.

NumPy-only Temporal Convolutional Network for gold volatility-regime forecasting using Bloomberg market data and HAR-RV baselines.

HAR-RV volatility forecasting baseline with ML overlay — USDZAR/JSE Top 40

MSc thesis investigating whether rough volatility models improve out-of-sample volatility forecasting performance relative to HAR-RV

BSc thesis (8.1/10): an honest out-of-sample test of neural networks vs HAR-RV and GARCH for realized volatility forecasting across 8 equity indices.

A comprehensive volatility forecasting project using ARIMA-GARCH family models, statistical diagnostics, realized volatility analysis, OHLC-based realized volatility estimators and backtesting on financial time series data.