On the use of conditional main effects for de‐aliasing 2k−p designs

A 2017 JQT article presented a methodology for disentangling two‐factor interactions that are aliased with other two‐factor interactions in 2k−p${2^{k - p}}$ designs. The methodology is based on the concept of conditional main effects, and it resolves aliasing by considering active the confounded in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Quality and reliability engineering international 2022-04, Vol.38 (3), p.1157-1164
Hauptverfasser: Grima, Pere, Rodero, Lourdes, Tort‐Martorell, Xavier
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A 2017 JQT article presented a methodology for disentangling two‐factor interactions that are aliased with other two‐factor interactions in 2k−p${2^{k - p}}$ designs. The methodology is based on the concept of conditional main effects, and it resolves aliasing by considering active the confounded interaction whose estimated effect is similar to one of the main effects of the two factors that are involved in it. This paper shows that there is no justification for this choice and provides examples. The conclusion is that the proposed method may be useful for building parsimonious models, but in general, it is not appropriate for de‐aliasing confounded two‐factor interactions.
ISSN:0748-8017
1099-1638
DOI:10.1002/qre.2894