Abstract
A supersaturated design is a type of factorial design where the number of columns m is at least as large as the number of rows n, so n≤m. The use of these designs is helpful in conducting experiments where, from a large number of factors the experimenters need to screen out a few significant ones, while the run size is limited. Analyzing data in supersaturated designs is a difficult task, since the number of experimental runs is less than the number of factors to be examined. In this article we propose a method for analyzing data using a specific type of supersaturated designs. This method use heavily the special block-orthogonal structure of the supersaturated designs given by Tang and Wu (1995). As an example we use the block-stepwise method to show the improvement archived, in comparison to the stepwise regression, when he structure of the design is considered.
Keywords and phrases: