Answers are grounded in this paper and cite their sources.
Demonstrates improved statistical power for bivariate stochastic dominance tests, suggesting better performance for distribution comparisons.
Stochastic dominance is a pivotal concept for comparing distributions. Traditional tests for bivariate stochastic dominance rely on the fully nonparametric empirical estimator of the bivariate distribution function. However, these tests often exhibit low statistical power because the nonparametric estimator tends to have high variance, especially with small sample sizes. To improve this, we propose constructing a test statistic using semi-parametric copula-based estimators of the bivariate distribution. We select data-dependent critical values to control the probability of type I error and establish the consistency of the proposed test. Through numerical studies, we demonstrate that our method outperforms existing approaches. Additionally, the practical applicability of this approach is illustrated using a real-world data example.
No takes yet. Share an insight, caveat, or question.
Zhuang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: