Constructing c-Optimal Designs for Polynomial Regression without an Intercept

Research output: Contribution to journalArticlepeer-review

Abstract

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.
Translated title of the contributionПостроение c-оптимальных планов для полиномиальной регрессии без свободного члена.
Original languageEnglish
Pages (from-to)223–231
JournalVestnik St. Petersburg University: Mathematics
Volume53
Issue number2
StatePublished - 2020

Keywords

  • c-optimal designs
  • f '(z)-optimal designs
  • designs optimal for estimating the derivative
  • polynomial regression with no intercept

Fingerprint Dive into the research topics of 'Constructing c-Optimal Designs for Polynomial Regression without an Intercept'. Together they form a unique fingerprint.

Cite this