metaBMA: Bayesian Model Averaged Meta-Analysis with JZS Priors

A statistical software package for performing Bayesian model-averaged meta-analysis, as proposed by Gronau et al. (2017). It computes posterior model probabilities for null, fixed-, and random-effects models, allowing for a wide range of priors on effect size and heterogeneity. The tool, authored by Daniel W. Heck, supports models with continuous and discrete moderators using pre-compiled Stan models.

Use Cases

• Compare fixed- and random-effects meta-analysis models based on posterior model probabilities.
• Estimate overall mean effect size using Bayesian model averaging as described by Gronau et al. (2017).
• Fit and test meta-regression models with continuous or discrete moderators using Jeffreys-Zellner-Siow (JZS) priors.
• Compute Bayes factors for model comparison in meta-analytic contexts.

Strengths

• Implements a specific Bayesian model-averaging method for meta-analysis from a 2017 peer-reviewed publication.
• Allows user-defined priors for effect size and heterogeneity, offering flexibility in analysis.
• Supports analysis with moderators using pre-compiled Stan models for computational efficiency.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

Daniel W. Heck

Collection Method

Software package for statistical computation.