Statistical Inference for a Ratio of Dispersions Using Paired Samples

This is the authors' abstract. We don't add key points for this paper.

Wilcox (1990) examined the Type I and Type II error rates for several robust tests of H 0 : σ 1 2 /σ 2 2 = 1 in paired-data designs and concluded that a satisfactory solution does not yet exist. A confidence interval for a ratio of correlated mean absolute deviations is derived and performs well in small sample sizes across realistically nonnormal distributions. When used to test a hypothesis, the proposed confidence interval is almost as powerful as the most powerful test examined by Wilcox.