Robustness of bioequivalence procedures under Box-Cox alternatives

This study investigates the robustness of Schuirmann's two one-sided tests procedure under a set of Box-Cox alternatives. A simulation is conducted to study the level of significance and the power of the procedure. Empirical results show that Schuirmann's procedure is robust under Box-Cox alternatives when the mean of the underlying distribution of the reference formulation is known, which suggests that the assumptions of normality or lognormality are sufficient but may not be necessary. It appears that Schuirmann's procedure cannot maintain the proper level of significance when the mean of the reference formulation is unknown, even under the normality or lognormality assumptions.