Statistical Modeling for Experiments with Sliding Levels

Abstract

Design of experiment with related factors can be implemented by using the technique of sliding levels. Taguchi proposed an analysis strategy by re- centering and re-scaling the slide factors. Hamada and Wu (1995) showed by a counterexample that in many cases the interactions cannot be completely eliminated by Taguchi's strategy. They proposed an alternative method in which the slide factors are modeled by nested effects. In this work we show the inadequacy of either method when the objective of the experiment is response prediction. We propose an analysis method based on a response surface model, and demonstrate its superiority for prediction. We also study the relationships between these three modeling strategies. (joint work with Jeff Wu and Longcheen Huwang)