作者: P. Bauer , M. Budde
关键词:
摘要: In the comparison of various dose levels it can often be assumed that parameters to tested follow an order restriction. Two closed multiple test procedures for detecting highest level still providing a shift in response distribution as compared adjacent lower is proposed. One based on one sided comparisons between neighbouring doses, other uses Helmert-type contrast statistics. If sequence testing fixed advance suitably modified. The power simulated under assumption normally distributed responses constellations means. It with general Holm-type procedure discussed BUDDE & BAUER (1989).