Dear all,
I am doing a simulation for my model that works when I use only the rq()
command. However, since I need to use the varcov matrix for my Wald test, I need
to compute summary(rq(), cov=TRUE). But the simulation does not work because of
the error: tau + h > 1: error in summary.rq
I tried to use:
if (tau + h > 1)
stop("tau + h > 1: error in summary.rq")
But the Hall-Sheather bandwidth is very high because I also vary the number of
observations from 40 to 300.
Is there anyone that could help me?
Thanks in advance!
All the best,
Julia
[[alternative HTML version deleted]]