Transition models for change-point estimation in logistic regression

Roberto Pastor-Barriuso, Eliseo Guallar, Josef Coresh

Research output: Contribution to journalArticle

Abstract

Although a wide variety of change-point models are available for continuous outcomes, few models are available for dichotomous outcomes. This paper introduces transition methods for logistic regression models in which the dose-response relationship follows two different straight lines, which may intersect or may present a jump at an unknown change-point. In these models, the logit includes a differentiable transition function that provides parametric control of the sharpness of the transition at the change-point, allowing for abrupt changes or more gradual transitions between the two different linear trends, as well as for estimation of the location of the change-point. Linear-linear logistic models are particular cases of the proposed transition models. We present a modified iteratively reweighted least squares algorithm to estimate model parameters, and we provide inference procedures including a test for the existence of the change-point. These transition models are explored in a simulation study, and they are used to evaluate the existence of a change-point in the association between plasma glucose after an oral glucose tolerance test and mortality using data from the Mortality Follow-up of the Second National Health and Nutrition Examination Survey.

Original languageEnglish (US)
Pages (from-to)1141-1162
Number of pages22
JournalStatistics in Medicine
Volume22
Issue number7
DOIs
StatePublished - Apr 15 2003

Fingerprint

Change-point Estimation
Transition Model
Change Point
Logistic Regression
Logistic Models
Glucose
Mortality
Nutrition Surveys
Iteratively Reweighted Least Squares
Change-point Model
Glucose Tolerance Test
Linear Trend
Least-Squares Analysis
Logit
Dose-response
Nutrition
Logistic Regression Model
Logistic Model
Least Square Algorithm
Sharpness

Keywords

  • Change-point
  • Hypothesis test
  • Logistics regression
  • Segmented regression models
  • Statistical inference
  • Transition models

ASJC Scopus subject areas

  • Epidemiology

Cite this

Transition models for change-point estimation in logistic regression. / Pastor-Barriuso, Roberto; Guallar, Eliseo; Coresh, Josef.

In: Statistics in Medicine, Vol. 22, No. 7, 15.04.2003, p. 1141-1162.

Research output: Contribution to journalArticle

@article{07c74be79bb14c18abd129c226bd68ff,
title = "Transition models for change-point estimation in logistic regression",
abstract = "Although a wide variety of change-point models are available for continuous outcomes, few models are available for dichotomous outcomes. This paper introduces transition methods for logistic regression models in which the dose-response relationship follows two different straight lines, which may intersect or may present a jump at an unknown change-point. In these models, the logit includes a differentiable transition function that provides parametric control of the sharpness of the transition at the change-point, allowing for abrupt changes or more gradual transitions between the two different linear trends, as well as for estimation of the location of the change-point. Linear-linear logistic models are particular cases of the proposed transition models. We present a modified iteratively reweighted least squares algorithm to estimate model parameters, and we provide inference procedures including a test for the existence of the change-point. These transition models are explored in a simulation study, and they are used to evaluate the existence of a change-point in the association between plasma glucose after an oral glucose tolerance test and mortality using data from the Mortality Follow-up of the Second National Health and Nutrition Examination Survey.",
keywords = "Change-point, Hypothesis test, Logistics regression, Segmented regression models, Statistical inference, Transition models",
author = "Roberto Pastor-Barriuso and Eliseo Guallar and Josef Coresh",
year = "2003",
month = "4",
day = "15",
doi = "10.1002/sim.1045",
language = "English (US)",
volume = "22",
pages = "1141--1162",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

TY - JOUR

T1 - Transition models for change-point estimation in logistic regression

AU - Pastor-Barriuso, Roberto

AU - Guallar, Eliseo

AU - Coresh, Josef

PY - 2003/4/15

Y1 - 2003/4/15

N2 - Although a wide variety of change-point models are available for continuous outcomes, few models are available for dichotomous outcomes. This paper introduces transition methods for logistic regression models in which the dose-response relationship follows two different straight lines, which may intersect or may present a jump at an unknown change-point. In these models, the logit includes a differentiable transition function that provides parametric control of the sharpness of the transition at the change-point, allowing for abrupt changes or more gradual transitions between the two different linear trends, as well as for estimation of the location of the change-point. Linear-linear logistic models are particular cases of the proposed transition models. We present a modified iteratively reweighted least squares algorithm to estimate model parameters, and we provide inference procedures including a test for the existence of the change-point. These transition models are explored in a simulation study, and they are used to evaluate the existence of a change-point in the association between plasma glucose after an oral glucose tolerance test and mortality using data from the Mortality Follow-up of the Second National Health and Nutrition Examination Survey.

AB - Although a wide variety of change-point models are available for continuous outcomes, few models are available for dichotomous outcomes. This paper introduces transition methods for logistic regression models in which the dose-response relationship follows two different straight lines, which may intersect or may present a jump at an unknown change-point. In these models, the logit includes a differentiable transition function that provides parametric control of the sharpness of the transition at the change-point, allowing for abrupt changes or more gradual transitions between the two different linear trends, as well as for estimation of the location of the change-point. Linear-linear logistic models are particular cases of the proposed transition models. We present a modified iteratively reweighted least squares algorithm to estimate model parameters, and we provide inference procedures including a test for the existence of the change-point. These transition models are explored in a simulation study, and they are used to evaluate the existence of a change-point in the association between plasma glucose after an oral glucose tolerance test and mortality using data from the Mortality Follow-up of the Second National Health and Nutrition Examination Survey.

KW - Change-point

KW - Hypothesis test

KW - Logistics regression

KW - Segmented regression models

KW - Statistical inference

KW - Transition models

UR - http://www.scopus.com/inward/record.url?scp=0037446014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037446014&partnerID=8YFLogxK

U2 - 10.1002/sim.1045

DO - 10.1002/sim.1045

M3 - Article

VL - 22

SP - 1141

EP - 1162

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 7

ER -