Portfolio optimization using factor models
Abstract: In this thesis model predictive control (MPC) is used to dynamically optimize a portfolio where data is sampled at the closing price. Previous research has shown that MPC optimization applied on financial data can yield a portfolio that exceeds the value of traditional portfolio strategies. MPC has also been observed having computational advantages when return forecasts are updated when a new observation are sampled. Factor models such as the Capital As- set Pricing Model (CAPM) and Fama and French factor models are used to forecast the financial return of stocks taken from the Standard & Poor’s 500 index Global. Portfolio optimization are performed using single-period forecast where the portfolio contains one stock and a zero interest rate cash account and also a large portfolio with 10 stocks and a risk-free asset. Transactions cost are included to better reflect the real world and address prediction-error. The MPC portfolio are outperforming a buy and hold strategy in both risk and return. Between the factor models then difference is negligible in case of the small portfolio but both Fama and French models outperforms CAPM in the larger portfolio.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)