An algorithm to optimize portfolio weights for the first degree stochastic dominance

Unlike the mean-variance approach, the stochastic dominance approach is to form a portfolio that first-degree stochastically dominates a predetermined benchmark portfolio, e.g. KOSPI. Analytically defining the first derivative of the objective function, an optimal algorithm of nonlinear programming was developed to search a set of optimal weights systematically and tested with promising results against real data sets from Korean stock market.