On the fitting of generalized linear models with nonnegativity parameter constraints

J. W. McDonald, I. D. Diamond

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

We consider the problem of finding maximum likelihood estimates of a generalized linear model when some or all of the regression parameters are constrained to be nonnegative. The Kuhn-Tucker conditions of nonlinear programming can be used to characterize the solution of this constrained estimation problem when the maximum likelihood estimates exist and are unique. For the case of a generalized linear model with nonnegativity parameter constraints, the Kuhn-Tucker conditions are derived and utilized to provide a stopping rule for search algorithms for the constrained maximum likelihood estimates. Two examples are discussed.

Original languageEnglish
Pages (from-to)201-206
Number of pages6
JournalBiometrics
Volume46
Issue number1
DOIs
Publication statusPublished - 1990

Fingerprint

Likelihood Functions
Nonnegativity
Generalized Linear Model
Maximum Likelihood Estimate
Maximum likelihood
Kuhn-Tucker Conditions
Linear Models
linear models
Constrained Estimation
Stopping Rule
Nonlinear programming
Nonlinear Programming
Search Algorithm
Regression
Non-negative

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Immunology and Microbiology(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

On the fitting of generalized linear models with nonnegativity parameter constraints. / McDonald, J. W.; Diamond, I. D.

In: Biometrics, Vol. 46, No. 1, 1990, p. 201-206.

Research output: Contribution to journalArticle

McDonald, J. W. ; Diamond, I. D. / On the fitting of generalized linear models with nonnegativity parameter constraints. In: Biometrics. 1990 ; Vol. 46, No. 1. pp. 201-206.
@article{8b06831949d7475ea7d5132b0741fc52,
title = "On the fitting of generalized linear models with nonnegativity parameter constraints",
abstract = "We consider the problem of finding maximum likelihood estimates of a generalized linear model when some or all of the regression parameters are constrained to be nonnegative. The Kuhn-Tucker conditions of nonlinear programming can be used to characterize the solution of this constrained estimation problem when the maximum likelihood estimates exist and are unique. For the case of a generalized linear model with nonnegativity parameter constraints, the Kuhn-Tucker conditions are derived and utilized to provide a stopping rule for search algorithms for the constrained maximum likelihood estimates. Two examples are discussed.",
author = "McDonald, {J. W.} and Diamond, {I. D.}",
year = "1990",
doi = "10.2307/2531643",
language = "English",
volume = "46",
pages = "201--206",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - On the fitting of generalized linear models with nonnegativity parameter constraints

AU - McDonald, J. W.

AU - Diamond, I. D.

PY - 1990

Y1 - 1990

N2 - We consider the problem of finding maximum likelihood estimates of a generalized linear model when some or all of the regression parameters are constrained to be nonnegative. The Kuhn-Tucker conditions of nonlinear programming can be used to characterize the solution of this constrained estimation problem when the maximum likelihood estimates exist and are unique. For the case of a generalized linear model with nonnegativity parameter constraints, the Kuhn-Tucker conditions are derived and utilized to provide a stopping rule for search algorithms for the constrained maximum likelihood estimates. Two examples are discussed.

AB - We consider the problem of finding maximum likelihood estimates of a generalized linear model when some or all of the regression parameters are constrained to be nonnegative. The Kuhn-Tucker conditions of nonlinear programming can be used to characterize the solution of this constrained estimation problem when the maximum likelihood estimates exist and are unique. For the case of a generalized linear model with nonnegativity parameter constraints, the Kuhn-Tucker conditions are derived and utilized to provide a stopping rule for search algorithms for the constrained maximum likelihood estimates. Two examples are discussed.

UR - http://www.scopus.com/inward/record.url?scp=0025087899&partnerID=8YFLogxK

U2 - 10.2307/2531643

DO - 10.2307/2531643

M3 - Article

VL - 46

SP - 201

EP - 206

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -