On the impact of association measures in portfolio theory

This paper discusses the use of different association measures in portfolio problems. From among the association measures we highlight those that are consistent with the choices of risk-averse investors and we characterize semidefinite positive association measures. Additionally, we propose new portfolio selection problems that optimize the association between the portfolio and one or two market benchmarks. Finally, we discuss when, and how, we can use association measures to reduce the dimensionality of portfolio problems. An empirical analysis shows the impact of different association measures in portfolio selection problems and in portfolio reduction problems. It is document that although the proper usage of both a risk measure and an association measure can increase the performance of the portfolio, the impact of the latter is higher.