A Note onD-Optimal Designs for Partially Nonlinear Regression Models

1 February 1984

journal article
research article
Published by Taylor & Francis in Technometrics

Vol. 26 (1) , 59-61
https://doi.org/10.1080/00401706.1984.10487922

Abstract

Hill (1980) states that a D-optimal design for the estimation of a subset of the parameter vector in a partially nonlinear model depends on only those parameters that appear nonlinearly in the model. In this note we demonstrate that Hill's statement is incorrect and present a suffkient condition under which it is valid.

Keywords

This publication has 5 references indexed in Scilit:

D-Optimal Designs for Partially Nonlinear Regression Models
Technometrics, 1980
Design of Experiments for Subsets of Parameters
Technometrics, 1974
Experiments for Nonlinear Functions (R.A. Fisher Memorial Lecture)
Journal of the American Statistical Association, 1973
DESIGN OF EXPERIMENTS IN NON-LINEAR SITUATIONS
Biometrika, 1959
Locally Optimal Designs for Estimating Parameters
The Annals of Mathematical Statistics, 1953