Portfolio optimization in time of varying uncertainty

Abstract

In this thesis, detailed theoretical background on how one can optimize a portfolio based on modern portfolio theory will be discussed. Harry Markowitz introduced the modern portfolio theory in 1952, which can guide one on what to optimise for portfolio optimisation. Combined with convex quadratic programming, we can optimise a portfolio with n assets using computers. In the past two years, multiple events occurred leading towards in a more volatile asset pricing market. This change in volatility results in a more uncertain future, the goal of this thesis is therefore to research the movement of asset distribution within the optimal portfolio, in times of varying volatility. By answering this, we aim to understand how one can adapt to changes in uncertainty, concerning short-term portfolio management. By choosing a portfolio containing assets equipped with different volatility functions, we can simulate how different volatilities perform within a portfolio. Using Python to optimise a portfolio, we obtain a numerical solution for the optimisation problem. Optimising a portfolio over a period of time resulted in a moving asset distribution within the optimal portfolio in times of varying volatility. From this, we can conclude that the optimal portfolio tends to favour less risky assets.