Optimal designs for a class of nonlinear regression models
Holger Dette, Viatcheslav B. Melas, Andrey Pepelyshev
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 f0 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.
Create a lesson
Related papers
Energy-Guided Learning for Heat Equation with Noisy Source: Optimal Estimation and Generalization Bounds
Zixiang Zhou, Xiaohui Chen, Shanghua Teng
Recurrent Barycenter Encoder: Prediction, Simulation and Applications to Time Series Analysis
Javier I Almarza
Quantitative Stability and Sample Complexity for Regularized Gromov-Wasserstein Distance
Pengtao Li, Xiaohui Chen
Minimax Wasserstein Testing under Differential Privacy
Bonwoo Lee, Ilmun Kim, Cheolwoo Park
Lattice-Based Characterizations and Inference for Multivariate Dependence
Paramahansa Pramanik
On diagonal sections of Markov products of copulas
Roman Drnovšek, Damjana Kokol Bukovšek, Nik Stopar