Selecting the number of classes under latent class regression: a factor analytic analogue

Huang, Guan‐Hua

doi:10.1007/s11336-003-1083-3

Cited by 7 publications

(8 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to identifiability constraints and the ability to readily interpret parameter effects, it is prudent to keep the covariates x and z (or covariate sets x and z ) mutually exclusive [8, 46]. We suggest using the same covariate set ( x ) for both the initial latent class weights and transition probabilities model components because these two components both govern the assignment of latent classes to individuals across time.…”

Section: Covariate Extensions To Lctmsmentioning

confidence: 99%

Using group‐based latent class transition models to analyze chronic disability data from the National Long‐Term Care Survey 1984–2004

White

Erosheva

2013

Statistics in Medicine

View full text Add to dashboard Cite

Latent class transition models track how individuals move among latent classes through time, traditionally assuming a complete set of observations for each individual. In this paper, we develop group-based latent class transition models that allow for staggered entry and exit, common in surveys with rolling enrollment designs. Such models are conceptually similar to, but structurally distinct from, pattern mixture models of the missing data literature. We employ group-based latent class transition modeling to conduct an in-depth data analysis of recent trends in chronic disability among the U.S. elderly population. Using activities of daily living data from the National Long-Term Care Survey (NLTCS), 1982-2004, we estimate model parameters using the expectation-maximization algorithm, implemented in SAS PROC IML. Our findings indicate that declines in chronic disability prevalence, observed in the 1980s and 1990s, did not continue in the early 2000s as previous NLTCS cross-sectional analyses have indicated.

show abstract

Section: Covariate Extensions To Lctmsmentioning

confidence: 99%

Using group‐based latent class transition models to analyze chronic disability data from the National Long‐Term Care Survey 1984–2004

White

Erosheva

2013

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“…However, the orthogonality assumption holds to an increasingly close approximation as N → ∞ if x i and z im are independent (Huang, 2005). This assumption can be verified empirically by calculating the sample correlation matrix among covariates.…”

Section: Latent Class Membership Assignment When Incorporating Covarimentioning

confidence: 92%

“…To apply these algorithms to models (4) and (5), it is necessary to "eliminate" the covariate effects, and hence "marginalize" models (4) and (5). 592 PSYCHOMETRIKA This study adopts the marginalization process developed in Section 3.3.1 of Huang (2005). The strategy for achieving such marginalization can be motivated by the properties of added variable plots for linear regression models (Cook & Weisberg, 1982).…”

Section: Latent Class Membership Assignment When Incorporating Covarimentioning

confidence: 99%

“…The pseudo-residual (13) is defined by analogizing the iteratively reweighted least-squares of generalized linear models with the least-square estimates of linear regressions (Landwehr, Pregibon, & Shoemaker, 1984;Huang, 2005). Then, classify objects based on new response variables R im (continuous indicators) or R im (categorical indicators), as in the previous subsection.…”

Section: Latent Class Membership Assignment When Incorporating Covarimentioning

confidence: 99%

See 1 more Smart Citation

Optimization-Based Model Fitting for Latent Class and Latent Profile Analyses

2011

Self Cite

View full text Add to dashboard Cite

Statisticians typically estimate the parameters of latent class and latent profile models using the Expectation-Maximization algorithm. This paper proposes an alternative two-stage approach to model fitting. The first stage uses the modified k-means and hierarchical clustering algorithms to identify the latent classes that best satisfy the conditional independence assumption underlying the latent variable model. The second stage then uses mixture modeling treating the class membership as known. The proposed approach is theoretically justifiable, directly checks the conditional independence assumption, and converges much faster than the full likelihood approach when analyzing high-dimensional data. This paper also develops a new classification rule based on latent variable models. The proposed classification procedure reduces the dimensionality of measured data and explicitly recognizes the heterogeneous nature of the complex disease, which makes it perfect for analyzing high-throughput genomic data. Simulation studies and real data analysis demonstrate the advantages of the proposed method.

show abstract

“…factor analysis, cluster analysis), several recommendations exist for deciding on the number of classes (k) in categorical latent variable models (Nylund et al, 2007). However, validity of this approach is limited in cases when the expected frequencies for a large number of observed variable patterns are low (Huang, 2005). However, validity of this approach is limited in cases when the expected frequencies for a large number of observed variable patterns are low (Huang, 2005).…”

Section: Introductionmentioning

confidence: 99%

Class Evolution Tree

Kriston

Melchior²,

Hergert³

et al. 2011

International Journal of Rehabilitation Research

View full text Add to dashboard Cite

The aim of our study was to develop a graphical tool that can be used in addition to standard statistical criteria to support decisions on the number of classes in explorative categorical latent variable modeling for rehabilitation research. Data from two rehabilitation research projects were used. In the first study, a latent profile analysis was carried out in patients with cancer receiving an inpatient rehabilitation program to identify prototypical combinations of treatment elements. In the second study, growth mixture modeling was used to identify latent trajectory classes based on weekly symptom severity measurements during inpatient treatment of patients with mental disorders. A graphical tool, the Class Evolution Tree, was developed, and its central components were described. The Class Evolution Tree can be used in addition to statistical criteria to systematically address the issue of number of classes in explorative categorical latent variable modeling.

show abstract

Selecting the number of classes under latent class regression: a factor analytic analogue

Abstract: categorical data, factor analysis, finite mixture model, goodness of fit test, latent profile model, marginalization, residuals in generalized linear models, Monte Carlo simulation.,

Cited by 7 publications

References 32 publications

Using group‐based latent class transition models to analyze chronic disability data from the National Long‐Term Care Survey 1984–2004

Using group‐based latent class transition models to analyze chronic disability data from the National Long‐Term Care Survey 1984–2004

Optimization-Based Model Fitting for Latent Class and Latent Profile Analyses

Class Evolution Tree

Contact Info

Product

Resources

About