A penalized latent class model for ordinal data

Stacia M. Desantis, E. Andrés Houseman, Brent A. Coull, Anat Stemmer-Rachamimov, Rebecca Betensky

Research output: Contribution to journalArticle

Abstract

Latent class models provide a useful framework for clustering observations based on several features. Application of latent class methodology to correlated, high-dimensional ordinal data poses many challenges. Unconstrained analyses may not result in an estimable model. Thus, information contained in ordinal variables may not be fully exploited by researchers. We develop a penalized latent class model to facilitate analysis of high-dimensional ordinal data. By stabilizing maximum likelihood estimation, we are able to fit an ordinal latent class model that would otherwise not be identifiable without application of strict constraints. We illustrate our methodology in a study of schwannoma, a peripheral nerve sheath tumor, that included 3 clinical subtypes and 23 ordinal histological measures.

Original languageEnglish (US)
Pages (from-to)249-262
Number of pages14
JournalBiostatistics
Volume9
Issue number2
DOIs
StatePublished - Apr 1 2008

Fingerprint

Nerve Sheath Neoplasms
Latent Class Model
Ordinal Data
Neurilemmoma
Cluster Analysis
Research Personnel
High-dimensional Data
Ordinal Variables
Latent Class
Methodology
Nerve
Maximum Likelihood Estimation
Tumor
Clustering
Ordinal data
Latent class model
Model

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Statistics, Probability and Uncertainty

Cite this

Desantis, S. M., Houseman, E. A., Coull, B. A., Stemmer-Rachamimov, A., & Betensky, R. (2008). A penalized latent class model for ordinal data. Biostatistics, 9(2), 249-262. https://doi.org/10.1093/biostatistics/kxm026

A penalized latent class model for ordinal data. / Desantis, Stacia M.; Houseman, E. Andrés; Coull, Brent A.; Stemmer-Rachamimov, Anat; Betensky, Rebecca.

In: Biostatistics, Vol. 9, No. 2, 01.04.2008, p. 249-262.

Research output: Contribution to journalArticle

Desantis, SM, Houseman, EA, Coull, BA, Stemmer-Rachamimov, A & Betensky, R 2008, 'A penalized latent class model for ordinal data', Biostatistics, vol. 9, no. 2, pp. 249-262. https://doi.org/10.1093/biostatistics/kxm026
Desantis SM, Houseman EA, Coull BA, Stemmer-Rachamimov A, Betensky R. A penalized latent class model for ordinal data. Biostatistics. 2008 Apr 1;9(2):249-262. https://doi.org/10.1093/biostatistics/kxm026
Desantis, Stacia M. ; Houseman, E. Andrés ; Coull, Brent A. ; Stemmer-Rachamimov, Anat ; Betensky, Rebecca. / A penalized latent class model for ordinal data. In: Biostatistics. 2008 ; Vol. 9, No. 2. pp. 249-262.
@article{487b274d31ee4e01860d56e09996d11b,
title = "A penalized latent class model for ordinal data",
abstract = "Latent class models provide a useful framework for clustering observations based on several features. Application of latent class methodology to correlated, high-dimensional ordinal data poses many challenges. Unconstrained analyses may not result in an estimable model. Thus, information contained in ordinal variables may not be fully exploited by researchers. We develop a penalized latent class model to facilitate analysis of high-dimensional ordinal data. By stabilizing maximum likelihood estimation, we are able to fit an ordinal latent class model that would otherwise not be identifiable without application of strict constraints. We illustrate our methodology in a study of schwannoma, a peripheral nerve sheath tumor, that included 3 clinical subtypes and 23 ordinal histological measures.",
author = "Desantis, {Stacia M.} and Houseman, {E. Andr{\'e}s} and Coull, {Brent A.} and Anat Stemmer-Rachamimov and Rebecca Betensky",
year = "2008",
month = "4",
day = "1",
doi = "10.1093/biostatistics/kxm026",
language = "English (US)",
volume = "9",
pages = "249--262",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - A penalized latent class model for ordinal data

AU - Desantis, Stacia M.

AU - Houseman, E. Andrés

AU - Coull, Brent A.

AU - Stemmer-Rachamimov, Anat

AU - Betensky, Rebecca

PY - 2008/4/1

Y1 - 2008/4/1

N2 - Latent class models provide a useful framework for clustering observations based on several features. Application of latent class methodology to correlated, high-dimensional ordinal data poses many challenges. Unconstrained analyses may not result in an estimable model. Thus, information contained in ordinal variables may not be fully exploited by researchers. We develop a penalized latent class model to facilitate analysis of high-dimensional ordinal data. By stabilizing maximum likelihood estimation, we are able to fit an ordinal latent class model that would otherwise not be identifiable without application of strict constraints. We illustrate our methodology in a study of schwannoma, a peripheral nerve sheath tumor, that included 3 clinical subtypes and 23 ordinal histological measures.

AB - Latent class models provide a useful framework for clustering observations based on several features. Application of latent class methodology to correlated, high-dimensional ordinal data poses many challenges. Unconstrained analyses may not result in an estimable model. Thus, information contained in ordinal variables may not be fully exploited by researchers. We develop a penalized latent class model to facilitate analysis of high-dimensional ordinal data. By stabilizing maximum likelihood estimation, we are able to fit an ordinal latent class model that would otherwise not be identifiable without application of strict constraints. We illustrate our methodology in a study of schwannoma, a peripheral nerve sheath tumor, that included 3 clinical subtypes and 23 ordinal histological measures.

UR - http://www.scopus.com/inward/record.url?scp=41149173077&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=41149173077&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxm026

DO - 10.1093/biostatistics/kxm026

M3 - Article

C2 - 17626225

AN - SCOPUS:41149173077

VL - 9

SP - 249

EP - 262

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 2

ER -