reg_sandwich.pkg

d 'REG_SANDWICH': Stata module to compute cluster-robust (sandwich) variance estimators with small-sample corrections for linear regression
d
d 'reg_sandwich' provides cluster-robust variance estimators (i.e., sandwich
d estimators) for ordinary and weighted least squares linear regression models. 
d Several adjustments are incorporated to improve small-sample performance. The
d package includes functions for estimating linear regression models with
d cluster-robust variance-covariance matrices and for testing single- and
d multiple-contrast hypotheses based on Wald test statistics. Variance-covariance
d estimators are based on a version of the bias-reduced linearization estimator
d proposed by Bell and McCaffrey (2002) and further developed by Tipton and
d Pustejovsky (2015) and Pustejovsky and Tipton (2016). Tests of single regression
d coefficients use Satterthwaite corrections. Tests of multiple-contrast
d hypotheses use an approximation to Hotelling's T-squared distribution.
d 
d Distribution-Date: 20170530
d
F reg_sandwich.ado
F reg_sandwich.sthlp
F reg_sandwich_ttests.mo
F test_sandwich.ado
F test_sandwich.sthlp
F test_sandwich_ftests.mo