Diversified portfolios with different entropy measures

Abstract

One of the major issues for Markowitz mean–variance model is the errors in estimations cause “corner solutions” and low diversity in the portfolio. In this paper, we compare the mean–variance efficiency, realized portfolio values, and diversity of the models incorporating different entropy measures by applying multiple criteria method. Differing from previous studies, we evaluate twenty-three portfolio over-time rebalancing strategies with considering short-sales and various transaction costs in asset diversification. Using the data of the most liquid stocks in Taiwan’s market, our finding shows that the models with Yager’s entropy yield higher performance because they respond to the change in market by reallocating assets more effectively than those with Shannon’s entropy and with the minimax disparity model. Furthermore, including entropy in models enhances diversity of the portfolios and makes asset allocation more feasible than the models without incorporating entropy.