Antedependence models for nonstationary categorical longitudinal data with ignorable missingness: likelihood-based inference

Yunlong Xie; Dale L Zimmerman

doi:10.1002/sim.5763

Back

Antedependence models for nonstationary categorical longitudinal data with ignorable missingness: likelihood-based inference

Journal article

Peer reviewed

Antedependence models for nonstationary categorical longitudinal data with ignorable missingness: likelihood-based inference

Yunlong Xie and Dale L Zimmerman

Statistics in medicine, Vol.32(19), pp.3274-3289

08/30/2013

DOI: 10.1002/sim.5763

PMCID: PMC3885186

PMID: 23436682

View Online

Abstract

Time index-ordered random variables are said to be antedependent (AD) of order (p1 ,p2 , … ,pn ) if the kth variable, conditioned on the pk immediately preceding variables, is independent of all further preceding variables. Inferential methods associated with AD models are well developed for continuous (primarily normal) longitudinal data, but not for categorical longitudinal data. In this article, we develop likelihood-based inferential procedures for unstructured AD models for categorical longitudinal data. Specifically, we derive maximum likelihood estimators (MLEs) of model parameters; penalized likelihood criteria and likelihood ratio tests for determining the order of antedependence; and likelihood ratio tests for homogeneity across groups, time invariance of transition probabilities, and strict stationarity. We give closed-form expressions for MLEs and test statistics, which allow for the possibility of empty cells and monotone missing data, for all cases save strict stationarity. For data with an arbitrary missingness pattern, we derive an efficient restricted expectation-maximization algorithm for obtaining MLEs. We evaluate the performance of the tests by simulation. We apply the methods to longitudinal studies of toenail infection severity (measured on a binary scale) and Alzheimer's disease severity (measured on an ordinal scale). The analysis of the toenail infection severity data reveals interesting nonstationary behavior of the transition probabilities and indicates that an unstructured first-order AD model is superior to stationary and other structured first-order AD models that have previously been fit to these data. The analysis of the Alzheimer's severity data indicates that the antedependence is second order with time-invariant transition probabilities, suggesting the use of a second-order autoregressive cumulative logit model.

Algorithms

Alzheimer Disease - physiopathology

Anti-Infective Agents - therapeutic use

Cognition

Computer Simulation

Foot Dermatoses - drug therapy

Humans

Likelihood Functions

Longitudinal Studies - methods

Models, Statistical

Onychomycosis - drug therapy

Details

Title: Subtitle: Antedependence models for nonstationary categorical longitudinal data with ignorable missingness: likelihood-based inference
Creators: Yunlong Xie - National Institutes of Health
Dale L Zimmerman - University of Iowa
Resource Type: Journal article
Publication Details: Statistics in medicine, Vol.32(19), pp.3274-3289
DOI: 10.1002/sim.5763
PMID: 23436682
PMCID: PMC3885186
NLM abbreviation: Stat Med
ISSN: 0277-6715
eISSN: 1097-0258
Grant note: U01 AG016976 / NIA NIH HHS Z99 HD999999 / Intramural NIH HHS
Language: English
Date published: 08/30/2013
Academic Unit: Statistics and Actuarial Science; Biostatistics
Record Identifier: 9984257747702771

Metrics

21 Record Views

2 Times Cited - Web of Science