Resampling-based simultaneous confidence intervals to compare scale using deviances

Abstract

Resampling-based confidence intervals based on medians are examined for pairwise comparison of scale. Methods that have been found in the literature to be effective for comparing scale for two groups are extended to the case of all pairwise comparisons, using a Tukey-type adjustment to ensure simultaneous confidence. Coverage rate and interval width estimates were computed using simulated data. A permutation-based method was found to have the best coverage properties.