Analysis Methods for Unreplicated Factorial Experiments
The analysis of unreplicated designs concentrates much of interest, since these designs enable us to estimate the factorial effects using contrasts, while no degrees of freedom are left to estimate the error variance, so conventional ANOVA techniques cannot be applied to detect the active effects. In this paper we review two effective methods (Angelopoulos and Koukouvinos, J. Appl. Statist 35:277–281, 2008; Angelopoulos et al., Qual. Reliab. Eng. Int 26:223–233, 2010) for the identification of active factors in unreplicated experiments. An illustrative example of the application of the two methods is presented, as also a comparative simulation study, revealing the effectiveness of the two methods.
KeywordsTwo-level factorial designs Unreplicated experiments Outliers Projective property Power
- 1.Aboukalam, M.A.F.: Quick, easy and powerful analysis of unreplicated factorial designs. Commun.Statist.- Theory Meth. 34, 1169–1175 (2005)Google Scholar
- 4.Benski, H.C.: Use of a normality test to identify significant effects in factorial designs. J. Qual. Tech. 21, 174–178 (1989)Google Scholar
- 13.Montgomery, D.C.: Design and analysis of experiments, 6th edn. John Willey and Sons, New York (2009)Google Scholar
- 14.Voss, D.T., Wang, W.: Analysis of othogonal saturated designs. In: Screening Methods for Experimentation in Industry, Drug Discovery and Genetics, pp. 268–281. Springer, New York (2006)Google Scholar