Testing Normality in Designs With Many Parameters

Technometrics vol. 48 issue 3 - August 2006

Abstract: [This abstract is based on the authors' abstract.]Goodness-of-fit tests are considered for the assumption of normality of random errors in experimental designs where the variance of response varies with the levels of the covariates. The proposed tests permit fitting different mean functions at different levels of the covariate so that data collected from several similar experiments can be combined to improve the power of the test.

Keywords: Empirical model - Goodness of fit - Normality - Residuals - Design of experiments (DOE) - Design reviews