Bayesian and profile likelihood change point methods for modeling cognitive function over time

Charles B. Hall, Jun Ying, Lynn Kuo, Richard B. Lipton

Research output: Contribution to journalArticle

61 Citations (Scopus)

Abstract

Change point models are often used to model longitudinal data. To estimate the change point, Bayesian (Biometrika 62 (1975) 407; Appl. Statist. 41 (1992) 389; Biometrics 51 (1995) 236) or profile likelihood (Statist. Med. 19 (2000) 1555) methods may be used. We compare and contrast the two methods in analyzing longitudinal cognitive data from the Bronx Aging Study. The Bayesian method has advantages over the profile likelihood method in that it does not require all subjects to have the same change point. Caution must be taken regarding sensitivity to choice of prior distribution, identifiability, and goodness of fit. Analyses show that decline in memory precedes diagnosis of dementia by 7.5-8 years, and individual change points are not needed to model heterogeneity across subjects.

Original languageEnglish (US)
Pages (from-to)91-109
Number of pages19
JournalComputational Statistics and Data Analysis
Volume42
Issue number1-2
DOIs
StatePublished - Feb 19 2003

Fingerprint

Cognitive Modeling
Profile Likelihood
Change Point
Change-point Model
Dementia
Likelihood Methods
Identifiability
Bayesian Methods
Longitudinal Data
Biometrics
Goodness of fit
Prior distribution
Aging of materials
Data storage equipment
Model
Estimate
Cognitive modeling
Change point

Keywords

  • Bayesian analyses
  • Change points
  • Cognitive aging
  • Longitudinal data
  • Markov chain Monte Carlo
  • Mixed models

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Statistics, Probability and Uncertainty
  • Electrical and Electronic Engineering
  • Computational Mathematics
  • Numerical Analysis
  • Statistics and Probability

Cite this

Bayesian and profile likelihood change point methods for modeling cognitive function over time. / Hall, Charles B.; Ying, Jun; Kuo, Lynn; Lipton, Richard B.

In: Computational Statistics and Data Analysis, Vol. 42, No. 1-2, 19.02.2003, p. 91-109.

Research output: Contribution to journalArticle

@article{d91c675cbc914765ba4113a7947882e8,
title = "Bayesian and profile likelihood change point methods for modeling cognitive function over time",
abstract = "Change point models are often used to model longitudinal data. To estimate the change point, Bayesian (Biometrika 62 (1975) 407; Appl. Statist. 41 (1992) 389; Biometrics 51 (1995) 236) or profile likelihood (Statist. Med. 19 (2000) 1555) methods may be used. We compare and contrast the two methods in analyzing longitudinal cognitive data from the Bronx Aging Study. The Bayesian method has advantages over the profile likelihood method in that it does not require all subjects to have the same change point. Caution must be taken regarding sensitivity to choice of prior distribution, identifiability, and goodness of fit. Analyses show that decline in memory precedes diagnosis of dementia by 7.5-8 years, and individual change points are not needed to model heterogeneity across subjects.",
keywords = "Bayesian analyses, Change points, Cognitive aging, Longitudinal data, Markov chain Monte Carlo, Mixed models",
author = "Hall, {Charles B.} and Jun Ying and Lynn Kuo and Lipton, {Richard B.}",
year = "2003",
month = "2",
day = "19",
doi = "10.1016/S0167-9473(02)00148-2",
language = "English (US)",
volume = "42",
pages = "91--109",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "1-2",

}

TY - JOUR

T1 - Bayesian and profile likelihood change point methods for modeling cognitive function over time

AU - Hall, Charles B.

AU - Ying, Jun

AU - Kuo, Lynn

AU - Lipton, Richard B.

PY - 2003/2/19

Y1 - 2003/2/19

N2 - Change point models are often used to model longitudinal data. To estimate the change point, Bayesian (Biometrika 62 (1975) 407; Appl. Statist. 41 (1992) 389; Biometrics 51 (1995) 236) or profile likelihood (Statist. Med. 19 (2000) 1555) methods may be used. We compare and contrast the two methods in analyzing longitudinal cognitive data from the Bronx Aging Study. The Bayesian method has advantages over the profile likelihood method in that it does not require all subjects to have the same change point. Caution must be taken regarding sensitivity to choice of prior distribution, identifiability, and goodness of fit. Analyses show that decline in memory precedes diagnosis of dementia by 7.5-8 years, and individual change points are not needed to model heterogeneity across subjects.

AB - Change point models are often used to model longitudinal data. To estimate the change point, Bayesian (Biometrika 62 (1975) 407; Appl. Statist. 41 (1992) 389; Biometrics 51 (1995) 236) or profile likelihood (Statist. Med. 19 (2000) 1555) methods may be used. We compare and contrast the two methods in analyzing longitudinal cognitive data from the Bronx Aging Study. The Bayesian method has advantages over the profile likelihood method in that it does not require all subjects to have the same change point. Caution must be taken regarding sensitivity to choice of prior distribution, identifiability, and goodness of fit. Analyses show that decline in memory precedes diagnosis of dementia by 7.5-8 years, and individual change points are not needed to model heterogeneity across subjects.

KW - Bayesian analyses

KW - Change points

KW - Cognitive aging

KW - Longitudinal data

KW - Markov chain Monte Carlo

KW - Mixed models

UR - http://www.scopus.com/inward/record.url?scp=0242586115&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242586115&partnerID=8YFLogxK

U2 - 10.1016/S0167-9473(02)00148-2

DO - 10.1016/S0167-9473(02)00148-2

M3 - Article

VL - 42

SP - 91

EP - 109

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 1-2

ER -