Optimal Design in Nonlinear Multiresponse Estimation: Poisson Model for Filter Feeding

D-optimality criteria have been applied to construct locally optimal designs for a multiresponse, nonlinear model. Simulated annealing was used to perform the needed numerical optimization calculations, as this method can locate the global optimum of a function, and can efficiently handle constraint...

Full description

Saved in:
Bibliographic Details
Published in:Biometrics 1992-12, Vol.48 (4), p.1235-1248
Main Authors: Hatzis, Christos, Larntz, Kinley
Format: Article
Language:English
Subjects:
Biological and medical sciences
Biometrics
Design efficiency
Estimate reliability
Exact sciences and technology
Experiment design
Experimental design
Fundamental and applied biological sciences. Psychology
Invertebrates
Mathematics
Matrices
Parametric models
Probability and statistics
Protozoa
Sciences and techniques of general use
Simplex method
Simulated annealing
Statistical discrepancies
Statistics
The Consultant's Forum
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:D-optimality criteria have been applied to construct locally optimal designs for a multiresponse, nonlinear model. Simulated annealing was used to perform the needed numerical optimization calculations, as this method can locate the global optimum of a function, and can efficiently handle constraints in the independent variables. The calculated optimal designs greatly reduce variances of model parameter estimates, compared to variance from previously used empirical designs. The effect of several design variables, including the number of design points and the number of responses, on the efficiency of the design was investigated, and designs for various subsets of parameters were also calculated. New directions for the design of future experiments were suggested by this analysis.
ISSN:0006-341X
1541-0420
DOI:10.2307/2532715