Portfolio allocation using omega function: An empirical analysis

It is widely recognized that expected returns and covariances are not sufficient to characterize the statistical properties of securities in the context of portfolio selection. Therefore different models have been proposed. On one side the Markowitz model has been extended to higher moments and on the other side, starting from Sharpe ratio, a great attention has been addressed to the correct choice of the risk (or joint risk-performance) indicator. One such indicator has been proposed recently in the financial literature: the so-called Omega Function, that considers all the moments of the return distribution and whose properties are being investigated thoroughly. The main purpose of this paper is to investigate empirically, in an out-of-sample perspective, the portfolios obtained using higher moments and the Omega ratio. Moreover we analyze the impact of the target threshold (when the Omega Ratio is used) and the impact of different preferences for moments and comoments (when a highermoments approach is used) on portfolio allocation. Our empirical analysis is based on a portfolio composed of 12 Hedge fund indexes.