Resumen
Extensions are presented to the results of Davidson and Duclos (2007), whereby the nullhypothesis of restricted stochastic non dominance can be tested by both asymptotic andbootstrap tests, the latter having considerably better properties as regards both size andpower. In this paper, the methodology is extended to tests of higher-order stochastic dominance.It is seen that, unlike the first-order case, a numerical nonlinear optimisation problemhas to be solved in order to construct the bootstrap DGP. Conditions are provided for asolution to exist for this problem, and efficient numerical algorithms are laid out. The empiricallyimportant case in which the samples to be compared are correlated is also treated,both for first-order and for higher-order dominance. For all of these extensions, the bootstrapalgorithm is presented. Simulation experiments show that the bootstrap tests performconsiderably better than asymptotic tests, and yield reliable inference in moderately sizedsamples.