Premium
Analyzing unreplicated 2 k factorial designs by examining their projections into k ‐1 factors
Author(s) -
Angelopoulos P.,
Evangelaras H.,
Koukouvinos C.
Publication year - 2010
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.1048
Subject(s) - factorial experiment , factorial , fractional factorial design , design of experiments , mathematics , property (philosophy) , computer science , statistics , mathematical analysis , philosophy , epistemology
Unreplicated factorial designs are widely used as experimental designs because of the economy they offer in run size. However, they are difficult to analyze because there are no degrees of freedom left to estimate the experimental error. Many methods have been proposed for the analysis of such designs with Lenth's ( Technometrics 1989; 31 :469–473) and Dong's ( Statist. Sinica 1993; 3 :209–217) being the most popular. In this paper we take advantage of the projective property of factorial designs and we propose a simple yet effective method for analyzing unreplicated factorial designs. The results are compared through a simulation study with Lenth's and Dong's methods. Copyright © 2009 John Wiley & Sons, Ltd.