Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Maximum likelihood analysis of a general latent variable model with hierarchically mixed data.

Sik-Yum Lee1, Xin-Yuan Song

  • 1Department of Statistics, The Chinese University of Hong Kong, Hong Kong. sylee@sparc2.sta.cuhk.edu.hk

Biometrics
|September 2, 2004
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A new species of Laubuka (Cypriniformes: Danionidae) from the Sittaung River drainage, Myanmar.

Journal of fish biology·2026
Same author

A review of genus Laubuka (Cypriniformes: Danionidae) in Myanmar with description of two new species.

Zootaxa·2025
Same author

Development and validation of predictive models for distant metastasis and prognosis of gastroenteropancreatic neuroendocrine neoplasms.

Scientific reports·2025
Same author

The complex relationship between gut microbiota and Alzheimer's disease: A systematic review.

Ageing research reviews·2024
Same author

Molecular characteristics and systemic treatment options of liposarcoma: A systematic review.

Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie·2024
Same author

Soybean steroids improve crop abiotic stress tolerance and increase yield.

Plant biotechnology journal·2024

A novel two-level latent variable model framework is introduced for comparing complex submodels. This statistical method handles nonlinearities and mixed data types, offering robust model evaluation for diverse research.

Area of Science:

  • * Statistical modeling
  • * Psychometrics
  • * Data analysis

Background:

  • * Existing statistical models often struggle with complex hierarchical data structures.
  • * Analyzing nonlinear relationships and mixed data types (continuous, dichotomous, polytomous) simultaneously presents a significant challenge.
  • * A comprehensive framework for comparing various submodels within a unified structure is needed.

Purpose of the Study:

  • * To develop a general two-level latent variable model for comprehensive submodel comparison.
  • * To provide a flexible framework capable of analyzing nonlinear relationships and fixed covariates.
  • * To extend the methodology to accommodate hierarchically mixed continuous, dichotomous, and polytomous data.

Main Methods:

  • * Development of a general two-level latent variable model.

Related Experiment Videos

  • * Implementation of a Monte Carlo Expectation-Maximization (EM) algorithm for maximum likelihood estimation.
  • * Approximation of conditional expectations using Markov chain Monte Carlo (MCMC) simulation in the E-step.
  • * Conditional maximization in the M-step.
  • * Proposed procedure for calculating the observed-data log-likelihood and Bayesian Information Criterion (BIC).

    • Main Results:

    • * The developed framework successfully accommodates nonlinear latent variable relationships and fixed covariates at both measurement and structural levels.
    • * The methodology is demonstrated to be applicable to hierarchically mixed data types.
    • * The Monte Carlo EM algorithm effectively produces maximum likelihood estimates.
    • * The proposed procedure allows for effective model comparison using BIC.

    Conclusions:

    • * The proposed two-level latent variable model offers a powerful and flexible framework for complex statistical analyses and model comparison.
    • * The methodology effectively handles nonlinearities and mixed data types in hierarchical structures.
    • * The computational approach using MCMC and EM algorithm provides a viable method for estimation and model evaluation.