A mean-variance Portfolio Optimizing Trading Algorithm using regime-switching Economic Parameters

Abstract: In this master's thesis a model of algorithmic trading is constructed. The model aims to create an optimal investment portfolio consisting of a risk-free asset and a risky asset. The risky asset is in the form of a stock generated using regime-switching parameters with a Markov chain explaining the state of the economy. The optimization of the portfolio is carried out under certain assumptions and reasonable constraints on risk, transaction costs and amount traded. The constraint on nancial risk is implemented through the recognized mean-variance criterion, balancing the expected value of the portfolio against the variance of the portfolio after every time period. The algorithm is implemented using quadratic programming techniques in Matlab. By varying parameters of the model a sensitivity analysis is performed. Simulated scenarios and the behaviour of the algorithm is presented in graphs. The algorithm is found to be rational and outperforms a static portfolio in every scenario.