Peter Nystrup: The purpose of the project is to develop and test dynamic asset allocation strategies based on identification of regime shifts in financial time series using statistical models with time-varying parameters.
A better description of the non-stationary behavior of the data process is meant to ensure the robustness of the developed strategies out of sample. The strategies will span multiple asset classes through an approach based on identification and modeling of common risk factors.
Effective start/end date 15/11/2014 → 13/03/2018
Published as PhD report:
Dynamic Asset Allocation - Identifying Regime Shifts in Financial Time Series to Build Robust Portfolios
Supervisor: Henrik Madsen
Dynamical Systems section at DTU Compute