Computation of the Efficient Boundary in the E-S Portfolio Selection Model

Portfolio selection models based on expected value-semivariance (E-S) criteria have been suggested as offering certain advantages over the expected value-variance (E-V) approach. Although variance is more tractable mathematically, it has not always been satisfying to financial theorists (3, pp. 278–284, 5, 6, 7, pp. 193–194, and 10, pp. 72–73). In the pioneering work in portfolio analysis, Markowitz 7, p. 194 observed that semivariance concentrates on reducing losses as opposed to variance which considers extreme gains, as well as extreme losses, as undesirable. In the presence of nonsymmetrical probability distributions, this equal weighting of gains and losses may not adequately describe the alternative portfolios available to the decision maker.