This course analyses the data toolkit underlying various investment strategies adopted by modern portfolio managers to optimize the risk-return tradeoff in wealth management. Examples of such strategies include (i) Exchange Traded Products replicating cross-sectional risk-premia of smart exposures to particular risk factors, and (ii) dynamic asset portfolios that time predictable variations in returns. Using an intuitive teaching approach, we explain key data-analytic tools for portfolio management in a non technical way. We rely on interactive and online training sessions on a web platform to demonstrate their application in selected real-data problems. Main objectives of the course are to understand (i) properties of modern investment strategies, (ii) existing risk-return tradeoffs in financial markets and (iii) key methodologies to measure these tradeoffs.
We start by introducing several important financial data structures necessary to quantify risk-return tradeoffs in financial markets, together with the tools needed to measure these tradeoffs, working under the umbrella of Arbitrage Pricing Theory. We highlight informally some of the most important risk and risk premium factors detected in the literature, which are applied by modern professional investors to optimize risk and return of various investment strategies. Then, we focus on different (linear) models for (i) estimating return predictability structures and for (ii) identifying systematic risk-premium components in cross-sections of assets. We also address portfolio tilting methodologies creating target exposures to particular risk premia, as well as conceptual issues related to factor tradability and factor proliferation in recent financial research.