Cross posted on Stackoverflow.

Cross posted on Stackoverflow with a bounty of 200.

EDIT:

I think I have to clarify this question a little bit more. So what I am looking for, is a function in which I can provide both the vcov matrix (the vcov2sls), and have robust and clustered standard errors. However it seems that they both pertain to the same vcov matrix option. So if I supply one, I already have to make sure the se's are clustered and robust.. So I guess I am essentially asking how I can make the vcov2sls function have robust and clustered errors. Obviously any other solution leading to the same practical outcome would be great as well.

END OF EDIT

EDIT: I know that also lmtest:coeftest exists, but I want to able to use weights. See this link. I am having trouble figuring out if this is possible in lmtest:coeftest .