Analytical versus machine learning calibration: which one works best in portfolio optimization?

Files

Supervisors: Lassance, Nathan
Faculty: Louvain School of Management
Degree label: Master [120] : ingénieur de gestion, à finalité spécialisée
Abstract: This master's thesis conducts empirical research on the estimation via machine learning of the decision parameters of different financial portfolio optimization strategies. Using the cross-validation method, the aim of the dissertation is to see whether the out-of-sample performance of portfolios estimated via machine learning is superior to that of conventional, analytically estimated, with strong statistical assumptions, portfolios.