Robust Regression (rlm) with Newey West Standard Errors

Question

Since my data is influenced by outliers and autocorrelation as well as by heteroskedasticity, I try to build a robust regression. However, I am not sure if the function "rlm" from the MASS package is compatible with the Newey West standard errors. Does anyone know whether the combination of the two functions has an adverse effect on each other?

Here is a code example of what I am trying to accomplish:

fit1 <- rlm(wage ~ status + country + familystatus + region)
fit2 <- coeftest(fit1,vcov=NeweyWest(fit1, verbose=T))

I would be very pleased about a short feedback considering my issues.

Answer 1

According to the article of A. Zeileis, " Econometric Computing with HC and HAC Covariance Matrix Estimators ", the rlm function is compatible with the Newey West standard errors:

The HAC estimators are already available for generalized linear models (fitted by glm) and robust regression (fitted by rlm in package MASS).