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

A Penalized Likelihood Method for Structural Equation Modeling.

Po-Hsien Huang1,2, Hung Chen3, Li-Jen Weng4

  • 1Department of Psychology, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei, 10617, Taiwan.

Psychometrika
|April 19, 2017
PubMed
Summary

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

The efficiency of 0.01% topical atropine on the myopia prevention in low refractive error population with different refractive errors: a retrospective cohort study.

BMC ophthalmology·2026
Same author

NSUN2 mediates intestinal stem cell expansion and colorectal tumour initiation via MAPK/ERK signalling.

Cell death & disease·2026
Same author

A CD8αβ co-receptor modified to contain an intracellular CD28 signaling tail enhances TCR-engineered T cell function independent of solid-tumor-associated co-stimulatory ligands.

Nature communications·2026
Same author

Residual permutation tests for feature importance in machine learning.

The British journal of mathematical and statistical psychology·2025
Same author

Macrophage activation determines muscle wasting in pancreatic cancer.

Oncogene·2025
Same author

Transcriptome-wide RNA 5-methylcytosine profiles of human iPSCs and iPSC-derived cardiomyocytes.

Scientific data·2024

A new penalized likelihood (PL) method enhances structural equation modeling (SEM) by balancing model fit and complexity. This approach improves interpretability, especially with limited prior knowledge, aiding psychological research.

Area of Science:

  • Statistics
  • Psychometrics
  • Quantitative Psychology

Background:

  • Structural Equation Modeling (SEM) is widely used to explore relationships between observed and latent variables.
  • Traditional SEM methods can struggle with model complexity and interpretability, particularly when prior knowledge is limited.
  • Penalized Likelihood (PL) offers a novel approach to address these limitations in SEM.

Purpose of the Study:

  • To introduce and evaluate a penalized likelihood (PL) method for structural equation modeling (SEM).
  • To demonstrate how PL balances model goodness-of-fit with model complexity.
  • To highlight PL's utility in enhancing model interpretability and its application in psychological research.

Main Methods:

  • Developed a penalized likelihood (PL) method incorporating a penalty term to control SEM model complexity.
Keywords:
ECM algorithmfactor analysis modelmodel selectionoracle propertypenalized likelihoodstructural equation modeling

Related Experiment Videos

  • Utilized an expectation-conditional maximization algorithm for maximizing the PL criterion.
  • Derived asymptotic properties of the proposed PL method.

    • Main Results:

    • The PL method yields SEM models that effectively balance goodness-of-fit and complexity.
    • PL estimates are sparse, leading to enhanced interpretability of the final model.
    • Numerical experiments and real data illustrations confirmed the utility of the PL method.

    Conclusions:

    • The penalized likelihood (PL) method provides a valuable extension to traditional and exploratory SEM.
    • PL is particularly beneficial when substantive knowledge for model specification is limited.
    • This methodology enhances the interpretability and parsimony of SEM models in psychological research.