Multi-state models and missing covariate data: expectation–maximization algorithm for likelihood estimation

Wenjie Lou, Lijie Wan, Erin L. Abner, David W. Fardo, Hiroko Dodge, Richard J. Kryscio

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Multi-state models have been widely used to analyse longitudinal event history data obtained in medical and epidemiological studies. The tools and methods developed recently in this area require completely observed data. However, missing data within variables of interest are very common in practice, and they have been an issue in applications. We propose a type of expectation–maximization (EM) algorithm, which handles missingness within multiple binary covariates efficiently, for multi-state model applications. Simulation studies show that the EM algorithm performs well for both missing completely at random and missing at random covariate data. We apply the method to a longitudinal aging and cognition study data-set, the Klamath Exceptional Aging Project, whose data were collected at Oregon Health & Science University and integrated into the Statistical Models of Aging and Risk of Transition database at the University of Kentucky.

Original languageEnglish (US)
Pages (from-to)20-35
Number of pages16
JournalBiostatistics and Epidemiology
Volume1
Issue number1
DOIs
StatePublished - Jan 1 2017

Fingerprint

Life Change Events
Statistical Models
Cognition
Epidemiologic Studies
Databases
Health
Datasets

Keywords

  • EM algorithm
  • MAR
  • MCAR
  • missing covariates
  • Multi-state model

ASJC Scopus subject areas

  • Epidemiology
  • Health Informatics

Cite this

Multi-state models and missing covariate data : expectation–maximization algorithm for likelihood estimation. / Lou, Wenjie; Wan, Lijie; Abner, Erin L.; Fardo, David W.; Dodge, Hiroko; Kryscio, Richard J.

In: Biostatistics and Epidemiology, Vol. 1, No. 1, 01.01.2017, p. 20-35.

Research output: Contribution to journalArticle

Lou, Wenjie ; Wan, Lijie ; Abner, Erin L. ; Fardo, David W. ; Dodge, Hiroko ; Kryscio, Richard J. / Multi-state models and missing covariate data : expectation–maximization algorithm for likelihood estimation. In: Biostatistics and Epidemiology. 2017 ; Vol. 1, No. 1. pp. 20-35.
@article{f98f223ea41c42e79ef7a713bc29e591,
title = "Multi-state models and missing covariate data: expectation–maximization algorithm for likelihood estimation",
abstract = "Multi-state models have been widely used to analyse longitudinal event history data obtained in medical and epidemiological studies. The tools and methods developed recently in this area require completely observed data. However, missing data within variables of interest are very common in practice, and they have been an issue in applications. We propose a type of expectation–maximization (EM) algorithm, which handles missingness within multiple binary covariates efficiently, for multi-state model applications. Simulation studies show that the EM algorithm performs well for both missing completely at random and missing at random covariate data. We apply the method to a longitudinal aging and cognition study data-set, the Klamath Exceptional Aging Project, whose data were collected at Oregon Health & Science University and integrated into the Statistical Models of Aging and Risk of Transition database at the University of Kentucky.",
keywords = "EM algorithm, MAR, MCAR, missing covariates, Multi-state model",
author = "Wenjie Lou and Lijie Wan and Abner, {Erin L.} and Fardo, {David W.} and Hiroko Dodge and Kryscio, {Richard J.}",
year = "2017",
month = "1",
day = "1",
doi = "10.1080/24709360.2017.1306156",
language = "English (US)",
volume = "1",
pages = "20--35",
journal = "Biostatistics and Epidemiology",
issn = "2470-9360",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

TY - JOUR

T1 - Multi-state models and missing covariate data

T2 - expectation–maximization algorithm for likelihood estimation

AU - Lou, Wenjie

AU - Wan, Lijie

AU - Abner, Erin L.

AU - Fardo, David W.

AU - Dodge, Hiroko

AU - Kryscio, Richard J.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Multi-state models have been widely used to analyse longitudinal event history data obtained in medical and epidemiological studies. The tools and methods developed recently in this area require completely observed data. However, missing data within variables of interest are very common in practice, and they have been an issue in applications. We propose a type of expectation–maximization (EM) algorithm, which handles missingness within multiple binary covariates efficiently, for multi-state model applications. Simulation studies show that the EM algorithm performs well for both missing completely at random and missing at random covariate data. We apply the method to a longitudinal aging and cognition study data-set, the Klamath Exceptional Aging Project, whose data were collected at Oregon Health & Science University and integrated into the Statistical Models of Aging and Risk of Transition database at the University of Kentucky.

AB - Multi-state models have been widely used to analyse longitudinal event history data obtained in medical and epidemiological studies. The tools and methods developed recently in this area require completely observed data. However, missing data within variables of interest are very common in practice, and they have been an issue in applications. We propose a type of expectation–maximization (EM) algorithm, which handles missingness within multiple binary covariates efficiently, for multi-state model applications. Simulation studies show that the EM algorithm performs well for both missing completely at random and missing at random covariate data. We apply the method to a longitudinal aging and cognition study data-set, the Klamath Exceptional Aging Project, whose data were collected at Oregon Health & Science University and integrated into the Statistical Models of Aging and Risk of Transition database at the University of Kentucky.

KW - EM algorithm

KW - MAR

KW - MCAR

KW - missing covariates

KW - Multi-state model

UR - http://www.scopus.com/inward/record.url?scp=85054780653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054780653&partnerID=8YFLogxK

U2 - 10.1080/24709360.2017.1306156

DO - 10.1080/24709360.2017.1306156

M3 - Article

AN - SCOPUS:85054780653

VL - 1

SP - 20

EP - 35

JO - Biostatistics and Epidemiology

JF - Biostatistics and Epidemiology

SN - 2470-9360

IS - 1

ER -