This course explores the characteristics of financial time series, their stylized facts, and the importance of modeling them for risk management, asset management, and quantitative trading.
It presents the main classes of models proposed in the literature, as well as recent developments driven by machine learning and AI.
Students learn both parametric approaches (LSTMs, spectral methods, structural causal models, etc.) and semi-/non-parametric techniques (factor models, transformers, generative models, signatures, etc.) for forecasting returns and risk. Emphasis is placed on leveraging all available data through advanced ML architectures to capture temporal dependencies and market dynamics.
