Bootstrap Identification of Confidence Intervals for the Non-Linear DoE Model

Abstract

Design of experiment (DoE) is a methodology widely used in an industry and an academia. However the fundamentals of DoE are well known since first articles of R.A. Fisher, the uncertainty estimation is still the investigated issue due to the fact that non-linear outcome functions do not preserve the normal distribution. The analytical solutions are known only for a very limited number of transformation. Authors propose to involve a bootstrap approach to estimate the outcome uncertainty of the response surface model.

Bootstrap-based model selection in subset polynomial regression

“…Kleiner et al [14] used Bootstrap for massive data. Jacek et al [15] used the Bootstrap approach to estimate the uncertainty of surface response models. Chen et al [16] used a bootstrap analysis to measure individual and regional differences in relative concentrations of gammaaminobutyric acid in the human brain.…”