Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Apr 12, 2011
In this video, I demonstrate how to get R to produce robust standard errors without having to create the robust variance-covariance matrix yourself every time you do it (using either hccm() in car or vcovHC in sandwich()). The key is to use a command that extends summary.lm(), which I have renamed summaryR().
I also demonstrate how to conveniently use the robust variance-covariance matrix when conducting a linear hypothesis test, merely by using the white.adjust option to linearHypothesis. These two commands are quite useful if you want to use robust standard errors.
Some information on this video (including code that will allow you to install the summaryR() command) is available at my econometrics blog: