The Annals of Statistics

Optimal Designs for a Class of Polynomials of Odd or Even Degree

H. Dette

Full-text: Open access

Abstract

In the class of polynomials of odd (or even) degree up to the order $2r - 1 (2r)$ optimal designs are determined which minimize a product of the variances of the estimates for the highest coefficients weighted with a prior $\gamma = (\gamma_1,\ldots,\gamma_r)$, where the numbers $\gamma_j$ correspond to the models of degree $2j - 1 (2j)$ for $j = 1,\ldots,r$. For a special class of priors, optimal designs of a very simple structure are calculated generalizing the $D_1$-optimal design for polynomial regression of degree $2r - 1 (2r)$. The support of these designs splits up in three sets and the masses of the optimal design at the support points of every set are all equal. The results are derived in a general context using the theory of canonical moments and continued fractions. Some applications are given to the $D$-optimal design problem for polynomial regression with vanishing coefficients of odd (or even) powers.

Article information

Source
Ann. Statist., Volume 20, Number 1 (1992), 238-259.

Dates
First available in Project Euclid: 12 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.aos/1176348520

Digital Object Identifier
doi:10.1214/aos/1176348520

Mathematical Reviews number (MathSciNet)
MR1150342

Zentralblatt MATH identifier
0745.62075

JSTOR
links.jstor.org

Subjects
Primary: 62K05: Optimal designs
Secondary: 62J05: Linear regression

Keywords
Polynomial regression $D$-optimal design $D_1$-optimal design model robust design canonical moments ultraspherical polynomials Chebyshev polynomials

Citation

Dette, H. Optimal Designs for a Class of Polynomials of Odd or Even Degree. Ann. Statist. 20 (1992), no. 1, 238--259. doi:10.1214/aos/1176348520. https://projecteuclid.org/euclid.aos/1176348520


Export citation