Tracking a benchmark index in portfolio optimization with two-stage mixed integer programming model

Abstract

The investors wish to achieve higher portfolio return than the benchmark index return at minimum tracking error (TE) in enhanced index tracking. This study aims to develop the optimal portfolio to track the benchmark sectorial index in Malaysia with two-stage mixed integer programming (MIP) model by minimizing the TE at the first stage followed by maximizing the portfolio mean return at the second stage. The data consists of Technology Index and the index components from Malaysia stock market. The results indicate that the two-stage MIP model gives higher mean return than the benchmark sectorial index at minimum TE. This study is significant because it helps to develop the optimal portfolio with two-stage MIP model to outperform the benchmark sectorial index without holding all index components.