2019/05/31 by Anna A. Igolkina, Meshcheryakov Georgy, Georgy Meshcheryakov +1 · 137 citations
Computer Science · Decision Sciences · Mathematics · #Advanced Statistical Modeling Techniques #Data Analysis with R #Latent variable #Multivariate statistics #Psychometric Methodologies and Testing #Python (programming language) #R package #Software package #Structural equation modeling #stat.AP
paper · pdf · doi:10.1080/10705511.2019.1704289
published in Structural Equation Modeling A Multidisciplinary Journal 27(6), 952-963 (Taylor & Francis)
openalex created_date 2019/05/29 · arxiv created 2019/09/06 · openalex publication_date 2020/02/18 · arxiv updated 2021/06/02 · openalex updated_date 2026/08/05
Structural equation modeling (SEM) is a multivariate statistical technique for estimating complex relationships between observed and latent variables. Although numerous SEM packages currently exist, they each have limitations and more importantly they are not free or open-source. The only package that is both free and open-source is lavaan. However, because this package is written in R, it is often difficult to integrate it with other programming language functionalities. This paper provides an overview of a new Python package called semopy that was specifically developed to overcome these limitations. Using illustrative examples we introduce the new package and then compare its performance in accuracy and execution time to lavaan.