Optimal designs based on the maximum quasi-likelihood estimator

Gang Shen, Seung Won Hyun, Weng Kee Wong

Research output: Contribution to journalArticle

Abstract

We use optimal design theory and construct locally optimal designs based on the maximum quasi-likelihood estimator (MqLE), which is derived under less stringent conditions than those required for the MLE method. We show that the proposed locally optimal designs are asymptotically as efficient as those based on the MLE when the error distribution is from an exponential family, and they perform just as well or better than optimal designs based on any other asymptotically linear unbiased estimators such as the least square estimator (LSE). In addition, we show current algorithms for finding optimal designs can be directly used to find optimal designs based on the MqLE. As an illustrative application, we construct a variety of locally optimal designs based on the MqLE for the 4-parameter logistic (4PL) model and study their robustness properties to misspecifications in the model using asymptotic relative efficiency. The results suggest that optimal designs based on the MqLE can be easily generated and they are quite robust to mis-specification in the probability distribution of the responses.

Original languageEnglish (US)
Pages (from-to)128-139
Number of pages12
JournalJournal of Statistical Planning and Inference
Volume178
DOIs
StatePublished - Nov 1 2016
Externally publishedYes

Fingerprint

Quasi-maximum Likelihood
Locally Optimal Design
Maximum likelihood
Estimator
Misspecification
Maximum likelihood estimation
Asymptotic Relative Efficiency
Asymptotically Linear
Logistic Model
Unbiased estimator
Exponential Family
Least Squares Estimator
Probability Distribution
Optimal design
Quasi-maximum likelihood estimator
Robustness
Probability distributions
Logistics
Specifications

Keywords

  • Approximate design
  • Design efficiency
  • Dose-finding study
  • Equivalence theorem
  • Heteroscedasticity
  • Maximum likelihood estimator

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

Optimal designs based on the maximum quasi-likelihood estimator. / Shen, Gang; Hyun, Seung Won; Wong, Weng Kee.

In: Journal of Statistical Planning and Inference, Vol. 178, 01.11.2016, p. 128-139.

Research output: Contribution to journalArticle

@article{790c5a59f07448e6abcfdf2f2e1f4ac0,
title = "Optimal designs based on the maximum quasi-likelihood estimator",
abstract = "We use optimal design theory and construct locally optimal designs based on the maximum quasi-likelihood estimator (MqLE), which is derived under less stringent conditions than those required for the MLE method. We show that the proposed locally optimal designs are asymptotically as efficient as those based on the MLE when the error distribution is from an exponential family, and they perform just as well or better than optimal designs based on any other asymptotically linear unbiased estimators such as the least square estimator (LSE). In addition, we show current algorithms for finding optimal designs can be directly used to find optimal designs based on the MqLE. As an illustrative application, we construct a variety of locally optimal designs based on the MqLE for the 4-parameter logistic (4PL) model and study their robustness properties to misspecifications in the model using asymptotic relative efficiency. The results suggest that optimal designs based on the MqLE can be easily generated and they are quite robust to mis-specification in the probability distribution of the responses.",
keywords = "Approximate design, Design efficiency, Dose-finding study, Equivalence theorem, Heteroscedasticity, Maximum likelihood estimator",
author = "Gang Shen and Hyun, {Seung Won} and Wong, {Weng Kee}",
year = "2016",
month = "11",
day = "1",
doi = "10.1016/j.jspi.2016.07.002",
language = "English (US)",
volume = "178",
pages = "128--139",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier",

}

TY - JOUR

T1 - Optimal designs based on the maximum quasi-likelihood estimator

AU - Shen, Gang

AU - Hyun, Seung Won

AU - Wong, Weng Kee

PY - 2016/11/1

Y1 - 2016/11/1

N2 - We use optimal design theory and construct locally optimal designs based on the maximum quasi-likelihood estimator (MqLE), which is derived under less stringent conditions than those required for the MLE method. We show that the proposed locally optimal designs are asymptotically as efficient as those based on the MLE when the error distribution is from an exponential family, and they perform just as well or better than optimal designs based on any other asymptotically linear unbiased estimators such as the least square estimator (LSE). In addition, we show current algorithms for finding optimal designs can be directly used to find optimal designs based on the MqLE. As an illustrative application, we construct a variety of locally optimal designs based on the MqLE for the 4-parameter logistic (4PL) model and study their robustness properties to misspecifications in the model using asymptotic relative efficiency. The results suggest that optimal designs based on the MqLE can be easily generated and they are quite robust to mis-specification in the probability distribution of the responses.

AB - We use optimal design theory and construct locally optimal designs based on the maximum quasi-likelihood estimator (MqLE), which is derived under less stringent conditions than those required for the MLE method. We show that the proposed locally optimal designs are asymptotically as efficient as those based on the MLE when the error distribution is from an exponential family, and they perform just as well or better than optimal designs based on any other asymptotically linear unbiased estimators such as the least square estimator (LSE). In addition, we show current algorithms for finding optimal designs can be directly used to find optimal designs based on the MqLE. As an illustrative application, we construct a variety of locally optimal designs based on the MqLE for the 4-parameter logistic (4PL) model and study their robustness properties to misspecifications in the model using asymptotic relative efficiency. The results suggest that optimal designs based on the MqLE can be easily generated and they are quite robust to mis-specification in the probability distribution of the responses.

KW - Approximate design

KW - Design efficiency

KW - Dose-finding study

KW - Equivalence theorem

KW - Heteroscedasticity

KW - Maximum likelihood estimator

UR - http://www.scopus.com/inward/record.url?scp=84979660142&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84979660142&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2016.07.002

DO - 10.1016/j.jspi.2016.07.002

M3 - Article

VL - 178

SP - 128

EP - 139

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

ER -