Constructing c-Optimal Designs for Polynomial Regression without an Intercept

In this paper, we consider the problem of constructing c -optimal designs for polynomial regression without an intercept. The special case of c = f  '( z ) (i.e., the vector of derivatives of the regression functions at some point z is selected as vector c ) is considered. The analytical result...

Full description

Saved in:
Bibliographic Details
Published in:Vestnik, St. Petersburg University. Mathematics St. Petersburg University. Mathematics, 2020-04, Vol.53 (2), p.223-231
Main Authors: Melas, V. B., Shpilev, P. V.
Format: Article
Language:English
Subjects:
Analysis
Exact solutions
Mathematical analysis
Mathematics
Mathematics and Statistics
Numerical methods
Polynomials
Regression analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we consider the problem of constructing c -optimal designs for polynomial regression without an intercept. The special case of c = f  '( z ) (i.e., the vector of derivatives of the regression functions at some point z is selected as vector c ) is considered. The analytical results available in the literature are briefly reviewed. An effective numerical method for finding f  '( z )-optimal designs in cases in which an analytical solution cannot be constructed is proposed.
ISSN:1063-4541
1934-7855
DOI:10.1134/S1063454120020120