Four rank-based procedures for profile analysis of repeated measure responses are
\ndiscussed in detail. All four offer the user a complete analysis including estimation
\nof regression coefficients, tests of general linear hypotheses, and confidence
\nprocedures. The fitting for each is based on minimizing a norm, hence, their
\ngeometry is similar to that of the traditional LS analysis. Two of the analyses are
\nmultivariate with theory not requiring assumptions on the covariance structure
\nof the repeated measures, while the other two are univariate analyses with theory
\nrequiring compound symmetry covariance structure. All of the analyses are easily
\ncomputed with existing R software. An example is discussed in some detail, including
\na sensitivity analysis. A Monte Carlo study investigates the validity and
\npower of the analyses over the normal and Cauchy distributions and a large family
\nof contaminated normal distributions and over two covariance structures. Generally
\nthe rank-based procedures were valid. In the normally distributed situations,
\nthe traditional LS analysis was more powerful, but by little; on the other hand,
\nall the rank-based analyses dominated LS over the other distributions. One of the
\nunivariate analyses (ATR) performed better than the others over the compound
\nsymmetric situations.<\/p>\n