Some robust design strategies for percentile estimation in binary response models

Download

Preview

Postscript Download (420Kb)

Description/Abstract

For the problem of percentile estimation of a quantal response curve, the authors determine multi-objective designs which are robust with respect to misspecifications of the model assumptions. They propose a maximin approach based on efficiencies and provide designs that are simultaneously efficient with respect to the particular choice of various parameter regions and link functions. Furthermore, the authors deal with the problems of designing model and percentile robust experiments and give various examples of such designs,
which are calculated numerically.