manMetaVAR: From Person-Specific Dynamics to Population Inference: A Computationally Efficient Meta-Analytic Structural Equation Modeling Framework for Integrating Complex Multilevel Dynamic Models

Anonymous

2026-04-10

Description

Research compendium for the manuscript From Person-Specific Dynamics to Population Inference: A Computationally Efficient Meta-Analytic Structural Equation Modeling Framework for Integrating Complex Multilevel Dynamic Models. https://doi.org/10.0000/0000000000

Acknowledgments

This research was supported by the Prevention and Methodology Training Program (PAMT), through a T32 training grant from the National Institute on Drug Abuse (NIDA; T32 DA017629; MPIs: J. Maggs and S. Lanza), as well as by grant UL1TR002014-06 from the National Center for Advancing Translational Sciences (NCATS).

Computations for this research were performed on the Pennsylvania State University’s Institute for Computational and Data Sciences’ Roar supercomputer using SLURM for job scheduling (Yoo et al., 2003), GNU Parallel to run the simulations in parallel (Tange, 2021), and Apptainer to ensure a reproducible software stack (Kurtzer et al., 2017, 2021). See .sim/README.md and the scripts in the .sim folder in the GitHub repository for more details on how the simulations were performed.

Installation

You can install manMetaVAR from GitHub with:

if (!require("pak")) install.packages("pak")
pak::pkg_install("jeksterslab/manMetaVAR")

See Containers for containerized versions of the package.

R Packages

Person-specific discrete-time and continuous-time vector autoregressive models for multiple individuals are available in the fitVARMxID package on the Comprehensive R Archive Network (CRAN) (https://CRAN.R-project.org/package=fitVARMxID). Documentation and examples can be found on the accompanying website (https://jeksterslab.github.io/fitVARMxID/). Meta-analytic synthesis of dynamic model estimates, including fixed-, random-, and mixed-effects multivariate meta-analysis models, is available in the metaDyn package on the Comprehensive R Archive Network (CRAN) (https://CRAN.R-project.org/package=metaDyn). Documentation and examples can be found on the accompanying website (https://jeksterslab.github.io/metaDyn/).

More Information

See GitHub Pages for package documentation.