anwendung von multiplen testprozeduren im linearen regressionsmodell

summary: this paper deals with the application of multiple test procedures in the linear regression model. the first part of the paper comprises the foundations of multiple test problems and an exact presentation of the so called closure principle. then it will be shown how to apply closed test procedures to elementary linear hypotheses. a number of simulation experiments were performed to compare different test procedures empirically. the results show the superiority of the sequentially rejective test procedures over conventional multiple tests like the bonferroni test or the scheffe test. finally the application of the closure principle to hierarchical test structures is demonstrated. in this context a generalized test structure is introduced, which contains some known test structures as special cases.;