A0513
Title: Maximum diversification versus minimum risk: Which is better
Authors: Pierpaolo Uberti - University of Milano-Bicocca (Italy) [presenting]
Maria-Laura Torrente - University of Genova, Dipartimento di Economia (Italy)
Abstract: In well-defined experimental settings, the out-of-sample performance of two asset allocation paradigms is evaluated: minimum risk and maximum diversification. The experiment is performed in an out-of-sample long-only framework, considering proportional transaction costs and different lengths of the estimation window and of the holding period. The strategies are compared in terms of numerical stability, return, Sharpe ratio and risk, measured through the same risk measures used for the calculation of the optimal allocation: variance of returns, mean absolute deviation, Value-at-Risk and expected shortfall. We show that the maximum diversification strategies are very competitive, if not better in general, than the risk minimization allocations. This result confirms well-known empirical findings of naive investment strategies that are difficult to beat in practice. Risk minimization strategies require very accurate forecasts of future returns to perform well; they are extremely numerically unstable and therefore, the implementation costs are high, significantly penalizing the performance. On the contrary, maximum diversification strategies are more stable and less affected by transaction costs. Furthermore, they do not need accurate predictions of future returns and are effective in controlling the risk of the investment.
