The effective sample size in Bayesian information criterion for level‐specific fixed and random‐effect selection in a two‐level nested model
British Journal of Mathematical and Statistical Psychology · 2024
DOI: 10.1111/bmsp.12327 ↗Kurz gefasst
Fachartikel in der Zeitschrift British Journal of Mathematical and Statistical Psychology, 2024. Verfasst von Sun-Joo Cho und 2 weiteren Autor:innen. Beteiligte Einrichtung: Vanderbilt University, Nashville, Tennessee, USA. Hauptthemen laut MEDLINE: Software. Die Arbeit nennt 36 Literaturangaben.
Abstract (Originalsprache)
Popular statistical software provides the Bayesian information criterion (BIC) for multi‐level models or linear mixed models. However, it has been observed that the combination of statistical literature and software documentation has led to discrepancies in the formulas of the BIC and uncertainties as to the proper use of the BIC in selecting a multi‐level model with respect to level‐specific fixed and random effects. These discrepancies and uncertainties result from different specifications of sample size in the BIC's penalty term for multi‐level models. In this study, we derive the BIC's penalty term for level‐specific fixed‐ and random‐effect selection in a two‐level nested design. In this new version of BIC, called , this penalty term is decomposed into two parts if the random‐effect variance–covariance matrix has full rank: (a) a term with the log of average sample size per cluster and (b) the total number of parameters times the log of the total number of clusters. Furthermore, we derive the new version of BIC, called , in the presence of redundant random effects. We show that the derived formulae, and , adhere to empirical values via numerical demonstration and that ( indicating either or ) is the best global selection criterion, as it performs at least as well as BIC with the total sample size and BIC with the number of clusters across various multi‐level conditions through a simulation study. In addition, the use of is illustrated with a textbook example dataset.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Vanderbilt University, Nashville, Tennessee, USA.
- Riverside Insights, Itasca, Illinois, USA.
ORCID-Profile: Sun-Joo Cho, Hao Wu, Matthew Naveiras
Zitieren
APA
Cho, S. J., Wu, H., & Naveiras, M. (2024). The effective sample size in Bayesian information criterion for level‐specific fixed and random‐effect selection in a two‐level nested model. British Journal of Mathematical and Statistical Psychology. https://doi.org/10.1111/bmsp.12327
Vancouver
Cho SJ, Wu H, Naveiras M. The effective sample size in Bayesian information criterion for level‐specific fixed and random‐effect selection in a two‐level nested model. British Journal of Mathematical and Statistical Psychology. 2024. doi:10.1111/bmsp.12327
BibTeX
@article{Cho2024,
author = {Cho, Sun-Joo and Wu, Hao and Naveiras, Matthew},
title = {The effective sample size in Bayesian information criterion for level‐specific fixed and random‐effect selection in a two‐level nested model},
journal = {British Journal of Mathematical and Statistical Psychology},
year = {2024},
doi = {10.1111/bmsp.12327}
}
EUVIMED