Optimal design and the model validity range

Valery V. Fedorov; Grace Montepiedra; Christopher J. Nachtsheim

doi:10.1016/s0378-3758(98)00033-0

Optimal design and the model validity range

Valery V. Fedorov, Grace Montepiedra, Christopher J. Nachtsheim

Supply Chain and Operations

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

7 Scopus citations

Abstract

A class of model-robust optimal designs, based on an extension of the standard optimality criteria to cases where there exist some prior information on the validity of a response function, is considered. Under this set-up, the concept of the "model validity range" is introduced and explored. A necessary condition for optimality is obtained for the determinant criterion (D-optimality in the classical case). A modified version of this criterion is proposed and discussed. The corresponding results provide upper and low bounds for the original problem and help to construct approximate solutions, when contamination is relatively small. Optimal designs for simple but commonly used regression models are obtained and studied.

Original language	English (US)
Pages (from-to)	215-227
Number of pages	13
Journal	Journal of Statistical Planning and Inference
Volume	72
Issue number	1-2
DOIs	https://doi.org/10.1016/s0378-3758(98)00033-0
State	Published - Sep 1 1998

Bibliographical note

Funding Information:
The authors are thankful to the referees and the associate editor for their suggestions, which significantly improved the presentation of this paper. Valery Fedorov’s research was supported by the Applied Mathematical Sciences Research Program, Office of Energy Research, U.S. Department of Energy under contract DE-AC05-84OR21400 with Lockheed Martin Energy Systems, Inc.

Keywords

62K05
First-order iterative algorithm
Linear regression
Mean squared error
Model validity* D -optimality
Model-robust design

Access

10.1016/s0378-3758(98)00033-0

OpenUrl availability

Full text

Cite this

@article{bad653adf25e47fca7e7c9b711291512,

title = "Optimal design and the model validity range",

abstract = "A class of model-robust optimal designs, based on an extension of the standard optimality criteria to cases where there exist some prior information on the validity of a response function, is considered. Under this set-up, the concept of the {"}model validity range{"} is introduced and explored. A necessary condition for optimality is obtained for the determinant criterion (D-optimality in the classical case). A modified version of this criterion is proposed and discussed. The corresponding results provide upper and low bounds for the original problem and help to construct approximate solutions, when contamination is relatively small. Optimal designs for simple but commonly used regression models are obtained and studied.",

keywords = "62K05, First-order iterative algorithm, Linear regression, Mean squared error, Model validity* D -optimality, Model-robust design",

author = "Fedorov, {Valery V.} and Grace Montepiedra and Nachtsheim, {Christopher J.}",

note = "Funding Information: The authors are thankful to the referees and the associate editor for their suggestions, which significantly improved the presentation of this paper. Valery Fedorov{\textquoteright}s research was supported by the Applied Mathematical Sciences Research Program, Office of Energy Research, U.S. Department of Energy under contract DE-AC05-84OR21400 with Lockheed Martin Energy Systems, Inc.",

year = "1998",

month = sep,

day = "1",

doi = "10.1016/s0378-3758(98)00033-0",

language = "English (US)",

volume = "72",

pages = "215--227",

journal = "Journal of Statistical Planning and Inference",

issn = "0378-3758",

publisher = "Elsevier",

number = "1-2",

}

TY - JOUR

T1 - Optimal design and the model validity range

AU - Fedorov, Valery V.

AU - Montepiedra, Grace

AU - Nachtsheim, Christopher J.

N1 - Funding Information: The authors are thankful to the referees and the associate editor for their suggestions, which significantly improved the presentation of this paper. Valery Fedorov’s research was supported by the Applied Mathematical Sciences Research Program, Office of Energy Research, U.S. Department of Energy under contract DE-AC05-84OR21400 with Lockheed Martin Energy Systems, Inc.

PY - 1998/9/1

Y1 - 1998/9/1

N2 - A class of model-robust optimal designs, based on an extension of the standard optimality criteria to cases where there exist some prior information on the validity of a response function, is considered. Under this set-up, the concept of the "model validity range" is introduced and explored. A necessary condition for optimality is obtained for the determinant criterion (D-optimality in the classical case). A modified version of this criterion is proposed and discussed. The corresponding results provide upper and low bounds for the original problem and help to construct approximate solutions, when contamination is relatively small. Optimal designs for simple but commonly used regression models are obtained and studied.

AB - A class of model-robust optimal designs, based on an extension of the standard optimality criteria to cases where there exist some prior information on the validity of a response function, is considered. Under this set-up, the concept of the "model validity range" is introduced and explored. A necessary condition for optimality is obtained for the determinant criterion (D-optimality in the classical case). A modified version of this criterion is proposed and discussed. The corresponding results provide upper and low bounds for the original problem and help to construct approximate solutions, when contamination is relatively small. Optimal designs for simple but commonly used regression models are obtained and studied.

KW - 62K05

KW - First-order iterative algorithm

KW - Linear regression

KW - Mean squared error

KW - Model validity D -optimality

KW - Model-robust design

UR - http://www.scopus.com/inward/record.url?scp=0000442467&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000442467&partnerID=8YFLogxK

U2 - 10.1016/s0378-3758(98)00033-0

DO - 10.1016/s0378-3758(98)00033-0

M3 - Article

AN - SCOPUS:0000442467

SN - 0378-3758

VL - 72

SP - 215

EP - 227

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

IS - 1-2

ER -

Optimal design and the model validity range

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this