Detection of unobserved heterogeneity with growth mixture models

Latent growth curve models as structural equation models are extensively discussedin various research fields (Duncan et al., 2006). Recent methodological and statisticalextension are focused on the consideration of unobserved heterogeneity in empiricaldata. Muth´en extended the classical structural...

Full description

Main Authors: Reinecke, Jost, Mariotti, Luca
Format: Artículo
Language: Español
Published: 2015
Online Access: http://revistas.ucr.ac.cr/index.php/matematica/article/view/1416
http://hdl.handle.net/10669/12944
id RepoKERWA12944
recordtype dspace
spelling RepoKERWA129442017-08-08T18:50:22Z Detection of unobserved heterogeneity with growth mixture models Detection of unobserved heterogeneity with growth mixture models Reinecke, Jost Mariotti, Luca Latent growth curve models as structural equation models are extensively discussedin various research fields (Duncan et al., 2006). Recent methodological and statisticalextension are focused on the consideration of unobserved heterogeneity in empiricaldata. Muth´en extended the classical structural equation approach by mixture components,i. e. categorical latent classes (Muth´en 2002, 2004, 2007).The paper will discuss applications of growth mixture models with data from oneof the first panel studies in Germany which explore deviant and delinquent behavior ofadolescents (Reinecke, 2006a, 2006b). Observed as well as unobserved heterogeneitywill be considered with growth mixture models using the program Mplus (Muth´en& Muth´en, 2006). Special attention is given to the distribution of the substantivedependent variables as a count measures (Poisson distribution, zero-inflated Poissondistribution, cf. Nagin, 1999). Different model specifications with respect to substantivequestions will also be emphasized.Keywords: Panel data, growth mixture models, heterogeneity, Poisson distribution. Los modelos latentes de curvas de crecimiento, como modelos de escuaciones estructurales,son ampliamente discutidos en varios campos de investigaci´on (Duncanet al., (2006)). Extensiones metodol´ogicas y estad´?sticas recientes se enfocan en laconsideraci´on de heterogeneidad no observada en datos emp´?ricos. Muth´en extendi´oel enfoque cl´asico de ecuaciones estructurales por componentes de mezcla, es decirclases latentes categ´oricas (Muth´en 2002, 2004, 2007).El art´?culo discute aplicaciones de modelos de crecimiento de mezcla con datosde uno de los primeros estudios de panel en Alemania, que explora comportamiento desviado y delinquivo de adolescentes (Reinecke, 2006a, 2006b). La heterogeneidadobservada y no observada ser´a considerada con modelos de crecimiento de mezclausando el programa Mplus (Muth´en & Muth´en, 2006). Se dar´a especial atenci´ona la distribuci´on de las variables sustantivas dependientes como medidas de conteo(distribuci´on de Poisson, distribuci´on cero-inflada de Poisson, cf. Nagin, 1999). Sedar´a ´enfasis tambi´en a diferentes especificaciones de modelos con respecto a cuestionesimportantes.Palabras clave: Datos de panel, modelos de mezclas de crecimiento, heterogeneidad,distribuci´on de Poisson. 2015-05-19T18:51:20Z 2015-05-19T18:51:20Z 2009-02-27 00:00:00 2015-05-19T18:51:20Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion http://revistas.ucr.ac.cr/index.php/matematica/article/view/1416 http://hdl.handle.net/10669/12944 10.15517/rmta.v16i1.1416 es Revista de Matemática: Teoría y Aplicaciones Vol. 16 Núm. 1 2009 16-29 application/pdf
institution Universidad de Costa Rica
collection Repositorio KERWA
language Español
description Latent growth curve models as structural equation models are extensively discussedin various research fields (Duncan et al., 2006). Recent methodological and statisticalextension are focused on the consideration of unobserved heterogeneity in empiricaldata. Muth´en extended the classical structural equation approach by mixture components,i. e. categorical latent classes (Muth´en 2002, 2004, 2007).The paper will discuss applications of growth mixture models with data from oneof the first panel studies in Germany which explore deviant and delinquent behavior ofadolescents (Reinecke, 2006a, 2006b). Observed as well as unobserved heterogeneitywill be considered with growth mixture models using the program Mplus (Muth´en& Muth´en, 2006). Special attention is given to the distribution of the substantivedependent variables as a count measures (Poisson distribution, zero-inflated Poissondistribution, cf. Nagin, 1999). Different model specifications with respect to substantivequestions will also be emphasized.Keywords: Panel data, growth mixture models, heterogeneity, Poisson distribution.
format Artículo
author Reinecke, Jost
Mariotti, Luca
spellingShingle Reinecke, Jost
Mariotti, Luca
Detection of unobserved heterogeneity with growth mixture models
author_sort Reinecke, Jost
title Detection of unobserved heterogeneity with growth mixture models
title_short Detection of unobserved heterogeneity with growth mixture models
title_full Detection of unobserved heterogeneity with growth mixture models
title_fullStr Detection of unobserved heterogeneity with growth mixture models
title_full_unstemmed Detection of unobserved heterogeneity with growth mixture models
title_sort detection of unobserved heterogeneity with growth mixture models
publishDate 2015
url http://revistas.ucr.ac.cr/index.php/matematica/article/view/1416
http://hdl.handle.net/10669/12944
_version_ 1681715772705996800
score 11.764614