🤖 AI Summary
This work addresses the absence of a unified and reproducible platform for Bayesian factor models, which has hindered fair comparisons and efficient implementation in large-scale, complex datasets. To bridge this gap, the study introduces a standardized, modular computational framework that enables seamless integration and direct comparison of diverse modern Bayesian factor models, including those with random effects. Built upon this framework, the authors develop and publicly release the factorverse R package, substantially enhancing model deployment efficiency and reproducibility. The proposed framework not only facilitates rigorous empirical evaluation across modeling approaches but also offers practical utility for both applied research and educational purposes.
📝 Abstract
Bayesian Factor Models (BFM) are well-established models that decompose the observed variability in a set of mean-zero, independent, and uncorrelated factors (random effects). While Factor Analysis (FA) was introduced in 1904 by Spearman, there has been renewed interest in inferential and computational methods that can adapt to large and complex modern data sets that are now routinely collected in a variety of applications. We provide reproducible, harmonized, and fast software for a variety of recent BFMs that allows the direct comparison of methods and provides a one-stop tutorial for the BFMs and their implementation. We neither endorse nor recommend any of the methods for a particular application; we simply provide a previously unavailable harmonized and reproducible common platform for BFMs. The accompanying factorverse R package is available at https://github.com/peterdunson/factorverse.