Designs for Selected Nonlinear Models

This chapter is an example-based guide to optimal design for nonlinear regression models. For clarity, we restrict ourselves to models with only one continuous explanatory variable. The classical theory presented in Section 14.2 also holds for multivariable models, whereas the results shown in Section 14.3 have been developed for single-variable models. In practice, designs for multivariable models are usually found numerically due to the increased complexity (see, e.g., Yu 2011, or Yang et al. 2013 for some recent developments on algorithms). Further information on algorithms for design search is given inChapters 13 and 21.