ILr-optimal designs for a hierarchically ordered system of regression models

ILr-optimal designs are described for a kind of hierarchically ordered system of regression models with an rr-dimensional response variable yy. The components of yy may be correlated with a known variance–covariance matrix ΣΣ. The present results show that ILr-optimal designs for this system of regression models do not depend on ΣΣ. The ILr-optimal designs are given for the systems of trigonometric and Haar wavelet regression models, respectively.