Document Detail


Bayesian model comparison of nonlinear structural equation models with missing continuous and ordinal categorical data.
MedLine Citation:
PMID:  15171804     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
Missing data are very common in behavioural and psychological research. In this paper, we develop a Bayesian approach in the context of a general nonlinear structural equation model with missing continuous and ordinal categorical data. In the development, the missing data are treated as latent quantities, and provision for the incompleteness of the data is made by a hybrid algorithm that combines the Gibbs sampler and the Metropolis-Hastings algorithm. We show by means of a simulation study that the Bayesian estimates are accurate. A Bayesian model comparison procedure based on the Bayes factor and path sampling is proposed. The required observations from the posterior distribution for computing the Bayes factor are simulated by the hybrid algorithm in Bayesian estimation. Our simulation results indicate that the correct model is selected more frequently when the incomplete records are used in the analysis than when they are ignored. The methodology is further illustrated with a real data set from a study concerned with an AIDS preventative intervention for Filipina sex workers.
Authors:
Sik-Yum Lee; Xin-Yuan Song
Related Documents :
16292114 - Methods for addressing missing data in psychiatric and developmental research.
9549814 - Statistical analysis of quality of life with missing data in cancer clinical trials.
11778674 - New approaches to missing data in psychological research: introduction to the special s...
18239164 - Empirical evidence of bias in the design of experimental stroke studies: a metaepidemio...
19767114 - A flock-tailored early warning system for low pathogenic avian influenza (lpai) in comm...
24833814 - Raman labeled nanoparticles: characterization of variability and improved method for un...
Publication Detail:
Type:  Comparative Study; Journal Article; Research Support, Non-U.S. Gov't    
Journal Detail:
Title:  The British journal of mathematical and statistical psychology     Volume:  57     ISSN:  0007-1102     ISO Abbreviation:  Br J Math Stat Psychol     Publication Date:  2004 May 
Date Detail:
Created Date:  2004-06-02     Completed Date:  2004-07-20     Revised Date:  2009-11-11    
Medline Journal Info:
Nlm Unique ID:  0004047     Medline TA:  Br J Math Stat Psychol     Country:  England    
Other Details:
Languages:  eng     Pagination:  131-50     Citation Subset:  IM    
Affiliation:
Department of Statistics, The Chinese University of Hong Kong.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms
Bayes Theorem*
Humans
Models, Theoretical*
Nonlinear Dynamics*
Psychology / statistics & numerical data*

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  The meta-analysis of partial effect sizes.
Next Document:  Three Mahalanobis distances and their role in assessing unidimensionality.