An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software

Gavin van der Nest; Valeria Lima Passos; Math J. J. M. Candel; Gerard J. P. van Breukelen

doi:10.1016/j.alcr.2019.100323

An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software

Gavin van der Nest^*, Valeria Lima Passos, Math J. J. M. Candel, Gerard J. P. van Breukelen

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

The use of finite mixture modelling (FMM) is becoming increasingly popular for the analysis of longitudinal repeated measures data. FMMs assist in identifying latent classes following similar paths of temporal development. This paper aims to address the confusion experienced by practitioners new to these methods by introducing the various available techniques, which includes an overview of their interrelatedness and applicability. Our focus will be on the commonly used model-based approaches which comprise latent class growth analysis (LCGA), group-based trajectory models (GBTM), and growth mixture modelling (GMM). We discuss criteria for model selection, highlight often encountered challenges and unresolved issues in model fitting, showcase model availability in software, and illustrate a model selection strategy using an applied example.

Original language	English
Article number	100323
Number of pages	17
Journal	Advances in Life Course Research
Volume	43
DOIs	https://doi.org/10.1016/j.alcr.2019.100323
Publication status	Published - Mar 2020

Keywords

Growth mixture model
Classification
Trajectory
Hidden heterogeneity
Latent class growth analysis
Repeated measures
CLASS GROWTH ANALYSIS
R PACKAGE
SAS PROCEDURE
MONTE-CARLO
2 DECADES
SELECTION
NUMBER
TRAJECTORIES
IMPACT
PERFORMANCE

Access to Document

10.1016/j.alcr.2019.100323Licence: CC BY

Cite this

@article{f11ff90a8111423dab1491ac898e56aa,

title = "An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software",

abstract = "The use of finite mixture modelling (FMM) is becoming increasingly popular for the analysis of longitudinal repeated measures data. FMMs assist in identifying latent classes following similar paths of temporal development. This paper aims to address the confusion experienced by practitioners new to these methods by introducing the various available techniques, which includes an overview of their interrelatedness and applicability. Our focus will be on the commonly used model-based approaches which comprise latent class growth analysis (LCGA), group-based trajectory models (GBTM), and growth mixture modelling (GMM). We discuss criteria for model selection, highlight often encountered challenges and unresolved issues in model fitting, showcase model availability in software, and illustrate a model selection strategy using an applied example.",

keywords = "Growth mixture model, Classification, Trajectory, Hidden heterogeneity, Latent class growth analysis, Repeated measures, CLASS GROWTH ANALYSIS, R PACKAGE, SAS PROCEDURE, MONTE-CARLO, 2 DECADES, SELECTION, NUMBER, TRAJECTORIES, IMPACT, PERFORMANCE",

author = "{van der Nest}, Gavin and Passos, {Valeria Lima} and Candel, {Math J. J. M.} and {van Breukelen}, {Gerard J. P.}",

note = "Publisher Copyright: {\textcopyright} 2020 The Authors",

year = "2020",

month = mar,

doi = "10.1016/j.alcr.2019.100323",

language = "English",

volume = "43",

journal = "Advances in Life Course Research",

issn = "1040-2608",

publisher = "JAI Press",

}

TY - JOUR

T1 - An overview of mixture modelling for latent evolutions in longitudinal data

T2 - Modelling approaches, fit statistics and software

AU - van der Nest, Gavin

AU - Passos, Valeria Lima

AU - Candel, Math J. J. M.

AU - van Breukelen, Gerard J. P.

PY - 2020/3

Y1 - 2020/3

N2 - The use of finite mixture modelling (FMM) is becoming increasingly popular for the analysis of longitudinal repeated measures data. FMMs assist in identifying latent classes following similar paths of temporal development. This paper aims to address the confusion experienced by practitioners new to these methods by introducing the various available techniques, which includes an overview of their interrelatedness and applicability. Our focus will be on the commonly used model-based approaches which comprise latent class growth analysis (LCGA), group-based trajectory models (GBTM), and growth mixture modelling (GMM). We discuss criteria for model selection, highlight often encountered challenges and unresolved issues in model fitting, showcase model availability in software, and illustrate a model selection strategy using an applied example.

AB - The use of finite mixture modelling (FMM) is becoming increasingly popular for the analysis of longitudinal repeated measures data. FMMs assist in identifying latent classes following similar paths of temporal development. This paper aims to address the confusion experienced by practitioners new to these methods by introducing the various available techniques, which includes an overview of their interrelatedness and applicability. Our focus will be on the commonly used model-based approaches which comprise latent class growth analysis (LCGA), group-based trajectory models (GBTM), and growth mixture modelling (GMM). We discuss criteria for model selection, highlight often encountered challenges and unresolved issues in model fitting, showcase model availability in software, and illustrate a model selection strategy using an applied example.

KW - Growth mixture model

KW - Classification

KW - Trajectory

KW - Hidden heterogeneity

KW - Latent class growth analysis

KW - Repeated measures

KW - CLASS GROWTH ANALYSIS

KW - R PACKAGE

KW - SAS PROCEDURE

KW - MONTE-CARLO

KW - 2 DECADES

KW - SELECTION

KW - NUMBER

KW - TRAJECTORIES

KW - IMPACT

KW - PERFORMANCE

U2 - 10.1016/j.alcr.2019.100323

DO - 10.1016/j.alcr.2019.100323

M3 - Article

SN - 1040-2608

VL - 43

JO - Advances in Life Course Research

JF - Advances in Life Course Research

M1 - 100323

ER -

An overview of mixture modelling for latent evolutions in longitudinal data: Modelling approaches, fit statistics and software

Abstract

Keywords

Access to Document

Fingerprint

Cite this