Factor selection in screening experiments by aggregation over random models

TL;DR

This paper introduces GDS-ARM, a novel method for screening experiments that aggregates results over multiple randomly selected interaction models to improve the identification of important factors.

Contribution

The paper proposes GDS-ARM, a new approach that combines Gauss-Dantzig Selector analyses on random interaction models to enhance factor screening accuracy.

Findings

GDS-ARM effectively identifies important factors in complex models.

The method performs well on both real and simulated data.

GDS-ARM reduces false positives compared to traditional methods.

Abstract

Screening experiments are useful for screening out a small number of truly important factors from a large number of potentially important factors. The Gauss-Dantzig Selector (GDS) is often the preferred analysis method for screening experiments. Just considering main-effects models can result in erroneous conclusions, but including interaction terms, even if restricted to two-factor interactions, increases the number of model terms dramatically and challenges the GDS analysis. We propose a new analysis method, called Gauss-Dantzig Selector Aggregation over Random Models (GDS-ARM), which performs a GDS analysis on multiple models that include only some randomly selected interactions. Results from these different analyses are then aggregated to identify the important factors. We discuss the proposed method, suggest choices for the tuning parameters, and study its performance on real and simulated data.