Confronting the ironies of optimal design: Nonoptimal sampling designs with desirable properties

1 March 1988

journal article
Published by American Geophysical Union (AGU) in Water Resources Research

Vol. 24 (3) , 409-415
https://doi.org/10.1029/wr024i003p00409

Abstract

Two sampling designs are developed for the improvement of parameter estimate precision in nonlinear regression, one for when there is uncertainty in the parameter values, and the other for when the correct model formulation is unknown. Although based on concepts of optimal design theory, the design criteria emphasize efficiency rather than optimality. The development is illustrated using a Streeter‐Phelps dissolved oxygen‐biochemical oxygen demand model.

Keywords

This publication has 15 references indexed in Scilit:

Experimental Design: Review and Comment
Technometrics, 1984
Optimum Composite Designs
Technometrics, 1974
On the Selection of Multi-Factor Experimental Arrangements with Resource Constraints
Journal of the American Statistical Association, 1971
On Practical Use of the Concept of D-Optimality
Technometrics, 1970
Optimal and Efficient Designs of Experiments
The Annals of Mathematical Statistics, 1969
The Design of Experiments for Parameter Estimation
Technometrics, 1968
Design of experiments for parameter estimation in multiresponse situations
Biometrika, 1966
DESIGN OF EXPERIMENTS IN NON-LINEAR SITUATIONS
Biometrika, 1959
On the Efficiency of Experimental Designs
The Annals of Mathematical Statistics, 1955
Optimum Allocation in Linear Regression Theory
The Annals of Mathematical Statistics, 1952