Optimum design accounting for the global nonlinear behavior of the model

DOI10.1214/14-AOS1232zbMATH Open1302.62174arXiv1310.5559MaRDI QIDQ464189FDOQ464189

Luc Pronzato, Andrej Pázman

Publication date: 17 October 2014

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: Among the major difficulties that one may encounter when estimating parameters in a nonlinear regression model are the nonuniqueness of the estimator, its instability with respect to small perturbations of the observations and the presence of local optimizers of the estimation criterion. We show that these estimability issues can be taken into account at the design stage, through the definition of suitable design criteria. Extensions of $E$-, $c$- and $G$-optimality criteria are considered, which when evaluated at a given $het{a}^0$ (local optimal design), account for the behavior of the model response $eta(heta)$ for $heta$ far from $het{a}^0$. In particular, they ensure some protection against close-to-overlapping situations where $|eta(heta)-eta(het{a}^0)|$ is small for some $heta$ far from $het{a}^0$. These extended criteria are concave and necessary and sufficient conditions for optimality (equivalence theorems) can be formulated. They are not differentiable, but when the design space is finite and the set $Theta$ of admissible $heta$ is discretized, optimal design forms a linear programming problem which can be solved directly or via relaxation when $Theta$ is just compact. Several examples are presented.

Full work available at URL: https://arxiv.org/abs/1310.5559

zbMATH Keywords

estimabilityoptimal designcurvaturenonlinear least-squares

Mathematics Subject Classification ID

General nonlinear regression (62J02) Optimal statistical designs (62K05)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Numerical optimization. Theoretical and practical aspects. Transl. from the French
• Introductory lectures on convex optimization. A basic course.
• The Equivalence of Two Extremum Problems
• Optimum Experimental Designs for Properties of a Compartmental Model
• Übersichtsarbeiten
• Orthogonal Array-Based Latin Hypercubes
• New variants of bundle methods
• The Cutting-Plane Method for Solving Convex Programs
• A Quadratic Design Criterion for Precise Estimation in Nonlinear Regression Models
• Necessary conditions for min-max problems and algorithms by a relaxation procedure
• Model-oriented design of experiments
• Optimal design for nonlinear response models
• Constrained design strategies for improving normal approximations in nonlinear regression problems
• Nonlinear experimental design based on the distribution of estimators
• Design of experiments in nonlinear models. Asymptotic normality, optimality criteria and small-sample properties
• Designs in nonlinear regression by stochastic minimization of functionals of the mean square error matrix
• On the regularization of singular c-optimal designs
• Nonlinear least squares - uniqueness versus ambiguity
• New Size $ \times $ Curvature Conditions for Strict Quasiconvexity of Sets
• Is this the least squares estimate?

Cited In (6)

• Design optimality for models defined by a system of ordinary differential equations
• An algorithm based on semidefinite programming for finding minimax optimal designs
• Optimal designs for nonlinear mixed-effects models using competitive swarm optimizer with mutated agents
• Optimal spatial monitoring of populations described by reaction-diffusion models
• The robustness against parameter variation of exact locally optimum designs in nonlinear regression -- a case study
• Optimal design of experiments via linear programming

Uses Software

• PLCP
• SQPlab

Recommendations

• Optimal design for nonlinear response models 👍 👎
• Global optimization approach to nonlinear optimal control 👍 👎
• Nonlinear multidisciplinary design optimization using system identification and optimal control theory 👍 👎
• Optimality criteria for design in nonlinear models with constraints 👍 👎
• Design optimality for models defined by a system of ordinary differential equations 👍 👎
• Optimal designs for some selected nonlinear models 👍 👎
• Title not available (Why is that?) 👍 👎
• Title not available (Why is that?) 👍 👎
• Title not available (Why is that?) 👍 👎

This page was built for publication: Optimum design accounting for the global nonlinear behavior of the model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q464189)