[R] Robust Standard Error in R

Achim Zeileis Achim.Zeileis at uibk.ac.at
Sun Sep 28 19:10:43 CEST 2014

On Sun, 28 Sep 2014, Arnab Dutta wrote:

> Hi,
>    In order to have robust standard errors in R, what would be the command
> that can generate results similar to the "robust" option in STATA?

This usually refers to sandwich standard errors aka HC or HC0 in case of 
the linear regression model. These are available in package "car" or 
package "sandwich". See vignette("sandwich", package = "sandwich") for a 
detailed overview.

> I tried using the "lmrob" command from the package "robustbase". With 
> that, the Adjusted R squared is quite different from the normal "lm" 
> command.

This performs robust estimation of the linear model while robust standard 
errors employ the usual OLS estimation and just adjust the subsequent 
inference. The robustness properties are quite different. While robust 
standard errors still require that the linear equation for the conditional 
mean holds for all observations, this is relaxed in robust estimates of 
the regression coefficients. More details can be found in the reference of 
the corresponding manuals.

> This does not happen in STATA. Can anybody please enlighten me on this?
> -Regards
> Arnab
> 	[[alternative HTML version deleted]]
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

More information about the R-help mailing list