Abstract: Supersaturated designs (SSDs) are useful in investigating a large number of factors with few experimental runs, particularly in screening experiments. The goal is to identify sparse but dominant active factors with small cost. In this paper, an analysis procedure called the Stepwise Response Refinement Screener (SRRS) method is proposed to screen important effects. Unlike the traditional approach that regresses factors with the response in every iteration, the response in each iteration of the SRRS is refined from the previous iteration using a selected potentially important factor. Analyses of two experiments using SSDs suggest that the SRRS method is able to retrieve similar results as do existing methods. Simulation studies show that, compared to existing methods in the literature, the SRRS method performs well in terms of the true model identification rate and the average model size.
Key words and phrases: Stepwise response refinement screener (SRRS), modified Akaike information criterion (mAIC), screening experiment, supersaturated design (SSD).