Rank-based tests for dose finding in nonmonotonic dose-response settings

Lim and Wolfe (1997, Biometrics 53, 410-418) proposed rank-based multiple t est procedures for identifying the dose levels that are more effective than the zero-dose control in randomized complete block designs when it can be assumed that the efficacy of the increasing dose levels is monotonically in creasing up to a point, followed by a monotonic decrease. Modifications of the Lim-Wolfe tests are suggested that provide more practical and powerful alternatives. Two numerical examples are illustrated and the results of a M onte Carlo power study are presented.