Tilting safety first and the Sharpe portfolio

“KLICing” there and back again: Portfolio selection using the empirical likelihood divergence and Hellinger distance

“…An application and a small-scale bootstrap study suggest the two new methods are competitive with and in some cases better than 9 We experimented with other values of d as well as other subsets of stocks; the result reported here are representative of our larger effort. 10 That the SR exhibits the smallest average shortfall probability is not necessarily surprising; Haley and McGee (2006) demonstrate that the SR can be interpreted as a disparity-based criterion that builds on the sum-of-squared deviations (Euclidean) metric; all four methods in Tables 3 and 4 are related in this sense.…”

“…4 Other disparity choices have been successfully explored in similar applications. For example, Robertson et al (2005) explore forecasting using different disparity measures (among them EL); Haley and McGee (2006) demonstrate that portfolio selection based on the sum-of-squared deviations (Euclidean) distance is the disparity-based representation of SF and the SR; and Haley and Walker (2010), building on Stutzer (1996) and Gray and Newman (2005), explore the merits of alternative disparity measures in option pricing. Additionally, the econometrics literature on minimum-disparity estimation makes extensive use of the EL divergence and HE distance.…”

L-performance with an application to hedge funds

“…The L-performance is based on the trimmed L-moments of order 1 and 2, which are more informative than the power moments, and exist even when the power moments do not. From an economic point of view, a Lperformance is an outcome of a portfolio optimizing behavior, based on alternative definitions of L-efficient portfolios and Lefficiency frontiers, in the spirit of Roy [Roy (1952), see also Bawa (1978), Haley and McGee (2006)]. The L-performance is parametrized by two (quantile) trimming parameters which allow for fine-tuning of L-performance according to financial risk criteria.…”

Portfolio performance evaluation with generalized Sharpe ratios: Beyond the mean and variance

“…These authors also show that the usage of the mean -lower partial moment objective corresponds to a specific utility function of the investor. Moreover, Haley and McGee (2006) demonstrated that such apparently various portfolio selection criteria as the Sharpe ratio, the safety first rule, and the Stutzer index can be obtained from a single behavioral assumption, namely that investors seek the portfolio that minimizes the probability of realizing a return below some pre-determined target or benchmark level. 3 In the literature there is also another approach to account for the skewness preferences, namely using the general equilibrium model.…”