A0311
Title: Some advances on test for stochastic dominance under density ratio model
Authors: Weiwei Zhuang - University of Science and Technology of China (China) [presenting]
Abstract: Stochastic dominance is a partial order between univariate random variables or their cumulative distribution functions. The stochastic dominance relationship is important in finance, economics and many other disciplines. Numerous approaches have been developed to test the hypotheses regarding stochastic dominance. The traditional estimation method is to use empirical distribution functions to estimate it. Considering the populations under comparison are generally of the same nature, we can link the populations through a density ratio model under certain condition. Based on this model, we propose some new estimators for stochastic dominance, restricted stochastic dominance and high order stochastic dominance. We improve the power of the tests through two venues: the use of resampling procedures and the confidence interval approach. Finally, we apply our method to analyze fund performance and family income.
