Nonlinear Structure Identification with Linear Least Squares and ANOVA

The objective of this paper is to find the structure of a nonlinear system from measurement data, as a prior step to model estimation. Applying ANOVA directly on a dataset is compared to applying ANOVA on residuals from a linear model. The distributions of the involved test variables are computed and used to show that ANOVA is effective in finding which regressors give linear effects and what regressors produce nonlinear effects. The ability to find nonlinear substructures depending on only subsets of regressors is an ANOVA feature which is shown not to be affected by subtracting a linear model