Detection of Interactions in Experiments on Large Numbers of Factors

Abstract

Summary. One of the main advantages of factorial experiments is the information that they can offer on interactions. When there are many factors to be studied, some or all of this information is often sacri®ced to keep the size of an experiment economically feasible. Two strategies for group screening are presented for a large number of factors, over two stages of experimentation, with particular emphasis on the detection of interactions. One approach estimates only main effects at the ®rst stage (classical group screening), whereas the other new method (interaction group screening) estimates both main effects and key two-factor interactions at the ®rst stage. Three criteria are used to guide the choice of screening technique, and also the size of the groups of factors for study in the ®rst-stage experiment. The criteria seek to minimize the expected total number of observations in the experiment, the probability that the size of the experiment exceeds a prespeci®ed target and the proportion of active individual factorial effects which are not detected. To implement these criteria, results are derived on the relationship between the grouped and individual factorial effects, and the probability distributions of the numbers of grouped factors whose main effects or interactions are declared active at the ®rst stage. Examples are used to illustrate the methodology, and some issues and open questions for the practical implementation of the results are discussed.

A comparison of design and model selection methods for supersaturated experiments

“…The Gauss-Dantzig selector has the advantages of having a robust automated procedure for the choice of δ, and a straightfoward interpretation of γ as the minimum size of an active effect. This quantity may often be elicited from subject experts (see, for example, ∆, in Lewis and Dean, 2001). 1.00 1.00 1.00 1.00 1.00 1.00 π 2 0.11 0.01 0.02 0.11 0.03 0.02 π 3…”

Design of Experiments for Screening

“…More recently, strategies for group screening that also investigate interactions at stage 1 have been developed [57]. In interaction group screening both main effects and two-variable interactions between the grouped variables are estimated at stage 1.…”

A Novel Methodology for Molecular Design via Data Driven Techniques

“…8 Factorial design technique appears to be very efficient for its possibility to systematically combine the levels being investigated for each complete trial or replication of the experiments. 9,10 In order to develop a property model, it is important to identify factors or independent variables that affect the response of interest or dependent variable. By utilising factorial design, effects and influence of the independent factors on the output of the dependent variable can be studied and investigated.…”