Size and Power of Some Stochastic Dominance Tests : A Monte Carlo Study

Testing for stochastic dominance between distributions is an important issue in the study of asset distribution, income distribution and market efficiency. This paper applies Monte Carlo simulations to examine the size and power of some commonly used stochastic dominance tests when the underlying distributions are correlated, and either homoskedastic or heteroskedastic. Our Monte Carlo study shows that the test suggested by Davidson and Duclos (2000) has a better size- and power-performance than two alternative tests developed by Kaur, Rao and Singh (1994) and Anderson (1996). In addition, when the underlying distributions are heteroskedastic, both the size and power of test developed by Davidson and Duclos (2000) are still superior to those of the two alternative tests