Optimal portfolio choice using the maximum Sharpe ratio
Abstract
Choosing a portfolio from among the enormous range of assets now available to an investor would be facilitated if we could locate the return–risk ratio of a particular allocation along a spectrum of possibilities. A comparison between portfolio choices can tell us, for example, whether it is better to select a suboptimal portfolio from a large class of assets or to perform a Markowitz optimal procedure on a subset of the assets. A common criterion for this assessment is the expected returntorisk tradeoff as measured by the Sharpe ratio. Given that the ideal, maximized Sharpe ratio must be estimated, we develop, in this paper, an approach that enables us to assess ex ante how close a given portfolio is to this ideal. For this purpose, we derive the largesample distribution of the maximized Sharpe ratio, as obtained from sample estimates, under very general assumptions. This distribution then represents the spectrum of possible optimal return–risk tradeoffs that can be constructed from thedata. We illustrate applications of the theory by analyzing a large sample of US companies, comparing constantcorrelation and momentum strategies with the optimal strategy. Simulations based on this data are also given for illustration.
Citation
Source Title
ISSN
Collection
Related items
Showing items related by title, author, creator and subject.

Pojanavatee, Sasipa (2013)Mutual funds are emerging as an opportunity for investors to automatically diversify their investments in such a way that all their money is pooled and the investment decisions are left to a professional manager. There ...

Maller, R.; Durand, Robert; Lee, P. (2005)We show that the maximum Sharpe ratio obtained via the Markowitz optimization procedure from a sample of returns on a number of risky assets is, under commonly satisfied assumptions, biased upwards for the population ...

Ling, A.; Sun, Jie; Yang, X. (2014)This paper proposes downside risk measure models in portfolio selection that captures uncertainties both in distribution and in parameters. The worstcase distribution with given information on the mean value and the ...