Application of Matlab in Portfolio Optimization

Abstract

Stock portfolios are estimated to diversify the risk in the financial market. When we invest in portfolios of stocks, we need to find the optimal portfolios with high expected return and low risk. Therefore, it is necessary to know which strategy performs the best and choose correct strategy to invest. The goal of this thesis is to apply different portfolio optimization strategies and to compare their out-of-sample results. In this thesis, we apply different strategies to calculate the weights of stock portfolios, and apply back testing method to obtain the returns and wealth of portfolios, then compare the performance of different strategies by Sharpe ratio and Maximum drawdown.In compliance with the results of all strategies that used, we make the ranking of the performance for these strategies, and choose which is the best strategy to invest.

Stock portfolios are estimated to diversify the risk in the financial market. When we invest in portfolios of stocks, we need to find the optimal portfolios with high expected return and low risk. Therefore, it is necessary to know which strategy performs the best and choose correct strategy to invest. The goal of this thesis is to apply different portfolio optimization strategies and to compare their out-of-sample results. In this thesis, we apply different strategies to calculate the weights of stock portfolios, and apply back testing method to obtain the returns and wealth of portfolios, then compare the performance of different strategies by Sharpe ratio and Maximum drawdown.In compliance with the results of all strategies that used, we make the ranking of the performance for these strategies, and choose which is the best strategy to invest.