Abstract

The paper deals with two issues. The first is testing the hypothesis that two dependent variables have a common variance. Currently, a heteroscedastic analog of the Morgan–Pitman test appears to perform relatively well. The paper demonstrates that when the marginal distributions differ in skewness, this method can be highly unsatisfactory. An explanation for this poor performance is provided. The second goal is to report simulation results on a percentile bootstrap method for comparing robust measures of scale. This approach was found to perform well in simulations, including the situation where the analog of the Morgan–Pitman test performs poorly.

摘要

这篇论文涉及两个问题。首先是检验两个因变量具有共同方差的假设。目前，Morgan-Pitman 检验的异方差类比似乎表现相对较好。该论文表明，当边际分布的偏度不同时，这种方法可能非常不令人满意。提供了对这种不良性能的解释。第二个目标是报告百分位自举方法的模拟结果，以比较稳健的规模测量。发现这种方法在模拟中表现良好，包括 Morgan-Pitman 测试的模拟表现不佳的情况。