umx

umx facilitates structural equation modeling (SEM) in R, enabling specification, estimation, and comparison of path-based and matrix-based models for social sciences and behavior genetics, including multigroup twin ACE and gene × environment models.


Key Features:

  • Path-Based Model Specification: Enables specification and estimation of path-based structural equation models for representing directed relationships among observed and latent variables.
  • Matrix-Based Models: Supports matrix-based model specification and estimation, including handling of raw and covariance data inputs and joint ordinal datasets.
  • Behavior Genetic Models: Implements standard multigroup twin models, including ACE (Additive genetic, Common environment, Unique environment) models, incorporation of covariates, and both common- and independent-pathway models.
  • Gene × Environment Interaction Models: Provides functionality for modeling gene × environment interaction within SEM frameworks.
  • Graphical and Tabular Outputs: Produces graphical and tabular results for interpretation and presentation of model estimates and comparisons.

Scientific Applications:

  • Social Sciences: Specification and testing of theoretical models with latent variables, path dependencies, and complex relationships among observed variables using SEM.
  • Behavior Genetics: Analysis of twin and multigroup genetic data to estimate heritability, shared and unique environmental influences, pathway structures, and gene × environment interactions.

Methodology:

Implements path-based and matrix-based SEM methods, including standard multigroup twin models (ACE and common/independent-pathway models), and supports analysis of raw, covariance, and joint ordinal datasets.

Topics

Details

License:
GPL-3.0
Maturity:
Mature
Cost:
Free of charge
Tool Type:
library
Operating Systems:
Linux, Windows, Mac
Programming Languages:
R
Added:
6/20/2019
Last Updated:
6/16/2020

Operations

Publications

Bates TC, Maes H, Neale MC. umx: Twin and Path-Based Structural Equation Modeling in R. Twin Research and Human Genetics. 2019;22(1):27-41. doi:10.1017/thg.2019.2. PMID:30944056.

Documentation

Links