Divisive latent class modeling as a density estimation method for categorical data

D.W. van der Palm, L.A. van der Ark, J.K. Vermunt

Research output: Contribution to journalArticleScientificpeer-review

Abstract

Traditionally latent class (LC) analysis is used by applied researchers as a tool for identifying substantively meaningful clusters. More recently, LC models have also been used as a density estimation tool for categorical variables. We introduce a divisive LC (DLC) model as a density estimation tool that may offer several advantages in comparison to a standard LC model. When using an LC model for density estimation, a considerable number of increasingly large LC models may have to be estimated before sufficient model-fit is achieved. A DLC model consists of a sequence of small LC models. Therefore, a DLC model can be estimated much faster and can easily utilize multiple processor cores, meaning that this model is more widely applicable and practical. In this study we describe the algorithm of fitting a DLC model, and discuss the various settings that indirectly influence the precision of a DLC model as a density estimation tool. These settings are illustrated using a synthetic data example, and the best performing algorithm is applied to a real-data example. The generated data example showed that, using specific decision rules, a DLC model is able to correctly model complex associations amongst categorical variables.

Original languageEnglish
Pages (from-to)52-72
JournalJournal of Classification
Volume33
Issue number1
DOIs
Publication statusPublished - 2016

Keywords

  • Latent class analysis
  • Categorical data
  • Mixture model
  • Density estimation
  • Divisive latent class model
  • Missing data
  • Multiple imputation
  • TEST-SCORE RELIABILITY
  • MULTIPLE IMPUTATION
  • COMPONENTS
  • INFERENCE
  • MIXTURES
  • NUMBER

Cite this

van der Palm, D.W. ; van der Ark, L.A. ; Vermunt, J.K. / Divisive latent class modeling as a density estimation method for categorical data. In: Journal of Classification. 2016 ; Vol. 33, No. 1. pp. 52-72.
@article{1cd738e05bce42b49a4a0bfde0b78873,
title = "Divisive latent class modeling as a density estimation method for categorical data",
abstract = "Traditionally latent class (LC) analysis is used by applied researchers as a tool for identifying substantively meaningful clusters. More recently, LC models have also been used as a density estimation tool for categorical variables. We introduce a divisive LC (DLC) model as a density estimation tool that may offer several advantages in comparison to a standard LC model. When using an LC model for density estimation, a considerable number of increasingly large LC models may have to be estimated before sufficient model-fit is achieved. A DLC model consists of a sequence of small LC models. Therefore, a DLC model can be estimated much faster and can easily utilize multiple processor cores, meaning that this model is more widely applicable and practical. In this study we describe the algorithm of fitting a DLC model, and discuss the various settings that indirectly influence the precision of a DLC model as a density estimation tool. These settings are illustrated using a synthetic data example, and the best performing algorithm is applied to a real-data example. The generated data example showed that, using specific decision rules, a DLC model is able to correctly model complex associations amongst categorical variables.",
keywords = "Latent class analysis, Categorical data, Mixture model, Density estimation, Divisive latent class model, Missing data, Multiple imputation, TEST-SCORE RELIABILITY, MULTIPLE IMPUTATION, COMPONENTS, INFERENCE, MIXTURES, NUMBER",
author = "{van der Palm}, D.W. and {van der Ark}, L.A. and J.K. Vermunt",
year = "2016",
doi = "10.1007/s00357-016-9195-5",
language = "English",
volume = "33",
pages = "52--72",
journal = "Journal of Classification",
issn = "0176-4268",
publisher = "Springer",
number = "1",

}

Divisive latent class modeling as a density estimation method for categorical data. / van der Palm, D.W.; van der Ark, L.A.; Vermunt, J.K.

In: Journal of Classification, Vol. 33, No. 1, 2016, p. 52-72.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Divisive latent class modeling as a density estimation method for categorical data

AU - van der Palm, D.W.

AU - van der Ark, L.A.

AU - Vermunt, J.K.

PY - 2016

Y1 - 2016

N2 - Traditionally latent class (LC) analysis is used by applied researchers as a tool for identifying substantively meaningful clusters. More recently, LC models have also been used as a density estimation tool for categorical variables. We introduce a divisive LC (DLC) model as a density estimation tool that may offer several advantages in comparison to a standard LC model. When using an LC model for density estimation, a considerable number of increasingly large LC models may have to be estimated before sufficient model-fit is achieved. A DLC model consists of a sequence of small LC models. Therefore, a DLC model can be estimated much faster and can easily utilize multiple processor cores, meaning that this model is more widely applicable and practical. In this study we describe the algorithm of fitting a DLC model, and discuss the various settings that indirectly influence the precision of a DLC model as a density estimation tool. These settings are illustrated using a synthetic data example, and the best performing algorithm is applied to a real-data example. The generated data example showed that, using specific decision rules, a DLC model is able to correctly model complex associations amongst categorical variables.

AB - Traditionally latent class (LC) analysis is used by applied researchers as a tool for identifying substantively meaningful clusters. More recently, LC models have also been used as a density estimation tool for categorical variables. We introduce a divisive LC (DLC) model as a density estimation tool that may offer several advantages in comparison to a standard LC model. When using an LC model for density estimation, a considerable number of increasingly large LC models may have to be estimated before sufficient model-fit is achieved. A DLC model consists of a sequence of small LC models. Therefore, a DLC model can be estimated much faster and can easily utilize multiple processor cores, meaning that this model is more widely applicable and practical. In this study we describe the algorithm of fitting a DLC model, and discuss the various settings that indirectly influence the precision of a DLC model as a density estimation tool. These settings are illustrated using a synthetic data example, and the best performing algorithm is applied to a real-data example. The generated data example showed that, using specific decision rules, a DLC model is able to correctly model complex associations amongst categorical variables.

KW - Latent class analysis

KW - Categorical data

KW - Mixture model

KW - Density estimation

KW - Divisive latent class model

KW - Missing data

KW - Multiple imputation

KW - TEST-SCORE RELIABILITY

KW - MULTIPLE IMPUTATION

KW - COMPONENTS

KW - INFERENCE

KW - MIXTURES

KW - NUMBER

U2 - 10.1007/s00357-016-9195-5

DO - 10.1007/s00357-016-9195-5

M3 - Article

VL - 33

SP - 52

EP - 72

JO - Journal of Classification

JF - Journal of Classification

SN - 0176-4268

IS - 1

ER -