We present and illustrate the methodology to calculate curvature measures for continuous designs, and extend design criteria to incorporate continuous designs. These design algorithms include quadratic design procedures, a subset design criterion, a second-order mean-square error design criterion, and a marginal curvature design methodology. A discussion of confidence intervals is also provided for continuous designs.
"Curvature measures for nonlinear regression models using continuous designs with applications to optimal experimental design." Involve 3 (3) 317 - 332, 2010. https://doi.org/10.2140/involve.2010.3.317