|
This article is cited in 2 scientific papers (total in 2 papers)
MATHEMATICS
Constructing c-optimal designs for polynomial regression with no intercept
V. B. Melas, P. V. Shpilev St. Petersburg State University, 7-9, Universitetskaya nab., St. Petersburg, 199034, Russian Federation
Abstract:
The paper is devoted to the problem of constructing c-optimal design for polynomial regression with no intercept. A special case of $c = f'(z)$ is considered (i. e., the vector of derivatives of regression functions at some point $z$ is selected as the vector $c$). A brief review of the analytical results are available in the literature is given. An effective numerical method for constructing $f'(z)$-optimal designs is proposed in those cases when an analytical solution cannot be provided.
Keywords:
c-optimal designs, $f'(z)$-optimal designs, optimal designs for estimating the slope, polynomial regression models with no intercept.
Received: 03.11.2019 Revised: 13.11.2019 Accepted: 12.12.2019
Citation:
V. B. Melas, P. V. Shpilev, “Constructing c-optimal designs for polynomial regression with no intercept”, Vestnik of Saint Petersburg University. Mathematics. Mechanics. Astronomy, 7:2 (2020), 331–342; Vestn. St. Petersbg. Univ., Math., 7:2 (2020), 223–231
Linking options:
https://www.mathnet.ru/eng/vspua194 https://www.mathnet.ru/eng/vspua/v7/i2/p331
|
|