I fit a quantile regression using quantreg:::rq on clustered data. I use the Huber sandwich estimator to obtain cluster-corrected standard errors, which is indicated by the se = 'nid' argument in summary.rq.

But here's my confusion:

q_1 <- rq(y ~ y, tau = .5, data = data)
summary.rq(q_1, se = 'nid')

Shouldn't there be an argument to specify on which variable is my data clustered? Thanks for the help!

*Below is a similar question to mine, but it doesn't resolve my specific confusion:

Quantile regression: Which standard errors?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.