Evolutionary multi-objective optimization in investment portfolio management

Abstract

This thesis aims to provide a comprehensive treatment on the design and application of multi-objective evolutionary algorithms to address several key issues involved with investment portfolio management. The central theme in investment portfolio management is the professional management of an appropriate mix of financial assets to satisfy specific investment goals. The decision process will typically involve issues such as asset allocation, security selection, performance measurement, management styles and etc. Due to the complexity of these issues, classical optimization tools from the realm of operations research are restricted to a limited set of problems and/or the optimization models have to accept strong simplifications. These issues, particularly asset allocation and management styles, were highlighted in this work and avenues to extend the generic MOEA platform for these purposes were proposed and empirically validated with datasets from actual equity indices in major financial markets.