When there are several endpoints and different predictors, researchers typically want to find out which predictors are relevant, and for which endpoints. We present two rather general approaches trying to accomplish these goals, accommodating different types (binary, ordinal, metric) of endpoints, and different designs. One of them uses rank-based statistics and an F-approximation of the sampling distribution, the other uses asymptotically valid resampling techniques. We also try to address the question of how well the proposed methods actually accomplish their goals.
Personal website of Arne Bathke