Experimental design for binary data

K. M. Abdelbasit, R. L. Plackett

Research output: Contribution to journalArticle

145 Citations (Scopus)

Abstract

Models for binary data are usually such that the information matrix depends on the unknown parameters. Thus the standard criteria for optimality in regression experiments cannot be applied without modification. Methods of going around this difficulty include the use of initial point estimates, sequential methods, and Bayesian analysis. This article is mainly concerned with the robustness and the number of design points for methods involving initial estimates, and for sequential methods in a small number of stages. A final section discusses the criterion of constant information for models involving one or two parameters, and summarizes recent results in this field.

Original languageEnglish
Pages (from-to)90-98
Number of pages9
JournalJournal of the American Statistical Association
Volume78
Issue number381
DOIs
Publication statusPublished - 1983

Fingerprint

Sequential Methods
Binary Data
Experimental design
Point Estimate
Information Matrix
Bayesian Analysis
Unknown Parameters
Two Parameters
Optimality
Regression
Robustness
Model
Estimate
Experiment
Design
Standards

Keywords

  • Binary data
  • Constant information
  • D-optimality
  • Fiducial method
  • Robustness
  • Sequential method

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Experimental design for binary data. / Abdelbasit, K. M.; Plackett, R. L.

In: Journal of the American Statistical Association, Vol. 78, No. 381, 1983, p. 90-98.

Research output: Contribution to journalArticle

@article{4fcff4b804e54180ad26d46361c059b5,
title = "Experimental design for binary data",
abstract = "Models for binary data are usually such that the information matrix depends on the unknown parameters. Thus the standard criteria for optimality in regression experiments cannot be applied without modification. Methods of going around this difficulty include the use of initial point estimates, sequential methods, and Bayesian analysis. This article is mainly concerned with the robustness and the number of design points for methods involving initial estimates, and for sequential methods in a small number of stages. A final section discusses the criterion of constant information for models involving one or two parameters, and summarizes recent results in this field.",
keywords = "Binary data, Constant information, D-optimality, Fiducial method, Robustness, Sequential method",
author = "Abdelbasit, {K. M.} and Plackett, {R. L.}",
year = "1983",
doi = "10.1080/01621459.1983.10477936",
language = "English",
volume = "78",
pages = "90--98",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "381",

}

TY - JOUR

T1 - Experimental design for binary data

AU - Abdelbasit, K. M.

AU - Plackett, R. L.

PY - 1983

Y1 - 1983

N2 - Models for binary data are usually such that the information matrix depends on the unknown parameters. Thus the standard criteria for optimality in regression experiments cannot be applied without modification. Methods of going around this difficulty include the use of initial point estimates, sequential methods, and Bayesian analysis. This article is mainly concerned with the robustness and the number of design points for methods involving initial estimates, and for sequential methods in a small number of stages. A final section discusses the criterion of constant information for models involving one or two parameters, and summarizes recent results in this field.

AB - Models for binary data are usually such that the information matrix depends on the unknown parameters. Thus the standard criteria for optimality in regression experiments cannot be applied without modification. Methods of going around this difficulty include the use of initial point estimates, sequential methods, and Bayesian analysis. This article is mainly concerned with the robustness and the number of design points for methods involving initial estimates, and for sequential methods in a small number of stages. A final section discusses the criterion of constant information for models involving one or two parameters, and summarizes recent results in this field.

KW - Binary data

KW - Constant information

KW - D-optimality

KW - Fiducial method

KW - Robustness

KW - Sequential method

UR - http://www.scopus.com/inward/record.url?scp=0002493307&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0002493307&partnerID=8YFLogxK

U2 - 10.1080/01621459.1983.10477936

DO - 10.1080/01621459.1983.10477936

M3 - Article

AN - SCOPUS:0002493307

VL - 78

SP - 90

EP - 98

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 381

ER -