[R-pkgs] new package clubSandwich: Cluster-Robust (Sandwich) Variance Estimators with Small-Sample Corrections

James Pustejovsky jepusto at gmail.com
Wed Jul 27 16:29:16 CEST 2016


Dear R users:

I'm happy to announce the first CRAN release of the clubSandwich package:

https://cran.r-project.org/web/packages/clubSandwich

clubSandwich provides several variants of the cluster-robust variance
estimator for ordinary and weighted least squares linear regression models,
including the bias-reduced linearization estimator of Bell and McCaffrey
(2002). The package includes functions for estimating the
variance-covariance matrix and for testing single- and multiple-contrast
hypotheses based on Wald test statistics. The hypothesis tests incorporate
small-sample corrections that lead to more accurate rejection rates when
the number of clusters is small or the design is unbalanced/leveraged.
Tests of single regression coefficients use Satterthwaite or saddle-point
corrections. Tests of multiple-contrast hypotheses use an approximation to
Hotelling's T-squared distribution. Methods are provided for a variety of
fitted models, including lm(), plm() (from package 'plm'), gls() and lme()
(from 'nlme'), robu() (from 'robumeta'), and rma.uni() and rma.mv() (from
'metafor').

The package includes two vignettes that demonstrate its use for estimation
of panel data models and meta-regression models.

Bug reports, suggestions, and feature requests are welcome
at https://github.com/jepusto/clubSandwich

Cheers,
James


___________________________________________
James Pustejovsky
Assistant Professor, Quantitative Methods Program
Educational Psychology Department
The University of Texas at Austin
http://jepusto.github.io/

	[[alternative HTML version deleted]]



More information about the R-packages mailing list