Optimal designs for a class of nonlinear regression models

Abstract

For a broad class of nonlinear regression models we investigate the local E- and c-optimal design problem. It is demonstrated that in many cases the optimal designs with respect to these optimality criteria are supported at the Chebyshev points, which are the local extrema of the equi-oscillating best approximation of the function f_0\equiv 0 by a normalized linear combination of the regression functions in the corresponding linearized model. The class of models includes rational, logistic and exponential models and for the rational regression models the E- and c-optimal design problem is solved explicitly in many cases.

Published at http://dx.doi.org/10.1214/009053604000000382 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (1)

• Optimal designs for a class of nonlinear regression models

Cited by in corpus (6)

• Optimal designs for a class of nonlinear regression models
• Approximate Dynamic Programming and Its Applications to the Design of Phase I Cancer Trials
• A geometric characterization of $c$-optimal designs for heteroscedastic regression
• The semi-algebraic geometry of saturated optimal designs for the Bradley-Terry model
• Incorporating Individual and Collective Ethics into Phase I Cancer Trial Designs
• Optimal designs for rational function regression