Latent variable models with ordinal categorical covariates
Publication in refereed journal

Times Cited
Web of Science5WOS source URL (as at 22/01/2021) Click here for the latest count
Altmetrics Information

Other information
AbstractWe propose a general latent variable model for multivariate ordinal categorical variables, in which both the responses and the covariates are ordinal, to assess the effect of the covariates on the responses and to model the covariance structure of the response variables. A fully Bayesian approach is employed to analyze the model. The Gibbs sampler is used to simulate the joint posterior distribution of the latent variables and the parameters, and the parameter expansion and reparameterization techniques are used to speed up the convergence procedure. The proposed model and method are demonstrated by simulation studies and a real data example.
All Author(s) ListPoon WY, Wang HB
Journal nameStatistics and Computing
Volume Number22
Issue Number5
PublisherSpringer Verlag (Germany)
Pages1135 - 1154
LanguagesEnglish-United Kingdom
KeywordsGibbs sampler; Latent variable model; Ordinal categorical variable; Parameter expansion; Reparameterization; Structural equation model
Web of Science Subject CategoriesComputer Science; Computer Science, Theory & Methods; COMPUTER SCIENCE, THEORY & METHODS; Mathematics; Statistics & Probability; STATISTICS & PROBABILITY

Last updated on 2021-23-01 at 01:48