mars R package
Software development project in R to create a unified meta-analytic and research synthesis analytic framework within R.
Open-Source mars package
The mars package includes functions for performing both univariate, multivariate, and multi-level meta-analysis, which involves synthesizing effect sizes from multiple primary studies into a single, more meaningful estimate. It also provides methods for estimating the asymptotic variance-covariance matrix of effect sizes. The optimization is done within the package directly and has implemented LASSO optimization routines to allow for robust estimation of meta-regression models with shrinkage.
Project Impact
The final product is an open-source R package, now available to the public. We’ve opened up a powerful meta-analytic method to a broader scientific community. Researchers can now use the package for univariate, multivariate, and multilevel analyses, with and without LASSO shrinkage.
Resources
The package can be installed from CRAN.
install.packages("mars")The full package website is available for you to view.