Standardized maximin criterion for discrimination and parameter estimation of nested models

Standard

Standardized maximin criterion for discrimination and parameter estimation of nested models. / Melas, Viatcheslav B. ; Guchenko , Roman ; Strashko, Vladislav .

In: Communications in Statistics: Simulation and Computation, Vol. 51, No. 8, 2022, p. 4314–4325.

Research output: Contribution to journal › Article › peer-review

BibTeX

@article{0bd7ae95c1f84d119fe8a468ba9082fb,

title = "Standardized maximin criterion for discrimination and parameter estimation of nested models",

abstract = "A new criterion for approximate designs called the standardized maximin criterion suited for both model discrimination and parameter estimation based on D- and D s-optimality criteria is introduced and studied. It is proved that the computation of an experimental design which is optimal with respect to this criterion can be reduced to the computation of multiple experimental designs which are optimal with respect to the simpler weighted criterion. Several numerical examples that describe the efficiency of the proposed criterion are provided. ",

keywords = "Experimental design, Maximin problems, Model discrimination, Parameter estimation, OPTIMUM DESIGNS, EQUIVALENCE, OPTIMAL EXPERIMENTAL-DESIGN",

author = "Melas, {Viatcheslav B.} and Roman Guchenko and Vladislav Strashko",

note = "Publisher Copyright: {\textcopyright} 2020 Taylor & Francis Group, LLC.",

year = "2022",

doi = "10.1080/03610918.2020.1741620",

language = "English",

volume = "51",

pages = "4314–4325",

journal = "Communications in Statistics Part B: Simulation and Computation",

issn = "0361-0918",

publisher = "Taylor & Francis",

number = "8",

}

RIS

TY - JOUR

T1 - Standardized maximin criterion for discrimination and parameter estimation of nested models

AU - Melas, Viatcheslav B.

AU - Guchenko , Roman

AU - Strashko, Vladislav

PY - 2022

Y1 - 2022

N2 - A new criterion for approximate designs called the standardized maximin criterion suited for both model discrimination and parameter estimation based on D- and D s-optimality criteria is introduced and studied. It is proved that the computation of an experimental design which is optimal with respect to this criterion can be reduced to the computation of multiple experimental designs which are optimal with respect to the simpler weighted criterion. Several numerical examples that describe the efficiency of the proposed criterion are provided.

AB - A new criterion for approximate designs called the standardized maximin criterion suited for both model discrimination and parameter estimation based on D- and D s-optimality criteria is introduced and studied. It is proved that the computation of an experimental design which is optimal with respect to this criterion can be reduced to the computation of multiple experimental designs which are optimal with respect to the simpler weighted criterion. Several numerical examples that describe the efficiency of the proposed criterion are provided.

KW - Experimental design

KW - Maximin problems

KW - Model discrimination

KW - Parameter estimation

KW - OPTIMUM DESIGNS

KW - EQUIVALENCE

KW - OPTIMAL EXPERIMENTAL-DESIGN

UR - http://www.scopus.com/inward/record.url?scp=85082338368&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/b4efc2d5-737d-3fdd-bebf-7c34f6f0ef3e/

U2 - 10.1080/03610918.2020.1741620

DO - 10.1080/03610918.2020.1741620

M3 - Article

VL - 51

SP - 4314

EP - 4325

JO - Communications in Statistics Part B: Simulation and Computation

JF - Communications in Statistics Part B: Simulation and Computation

SN - 0361-0918

IS - 8

ER -

ID: 52811346