REG_SANDWICH
: Stata module to compute cluster-robust (sandwich) variance estimators with small-sample corrections for linear regression
reg_sandwich
provides cluster-robust variance estimators (i.e., sandwich
estimators) for ordinary and weighted least squares linear regression models.
Several adjustments are incorporated to improve small-sample performance. (We like to think of these adjustments as extra cheese, sprouts, bacon, etc. in the middle of the sandwich estimator.) The package includes functions for estimating linear regression models with
cluster-robust variance-covariance matrices and for testing single- and
multiple-contrast hypotheses based on Wald test statistics. Variance-covariance
estimators are based on a version of the bias-reduced linearization estimator
proposed by Bell and McCaffrey (2002) and further developed by Tipton and
Pustejovsky (2015) and Pustejovsky and Tipton (2016). Tests of single regression
coefficients use Satterthwaite corrections. Tests of multiple-contrast
hypotheses use an approximation to Hotelling's T-squared distribution.
- Stata version 14.2
Marcelo Tyszler. Sustainable Economic Development and Gender, Royal Tropical Institute, Netherlands. [email protected]
James E. Pustejovsky (maintainer). Department of Education Psychology, University of Texas at Austin. [email protected]
Elizabeth Tipton. Department of Human Development, Teachers College, Columbia University. [email protected]
The package is available on the SSC Archive, under the name reg_sandwich
. To install it, type
ssc install reg_sandwich
To install the latest development version directly from Github, type:
net install github, from("https://haghish.github.io/github/")
github install jepusto/clubSandwich-Stata
Please cite the package as follows:
Tyszler, M., Pustejovsky, J. E., & Tipton, E. 2017. REG_SANDWICH: Stata module to compute cluster-robust (sandwich) variance estimators with small-sample corrections for linear regression, Statistical Software Components S458352, Boston College Department of Economics. URL: https://ideas.repec.org/c/boc/bocode/s458352.html