Skip to content

quantreg speed

2 messages · Roger Koenker, Yunqi Zhang

#
You could try method = "pin".  

Sent from my iPhone
#
Hi Roger,

Thank you for your reply. To my understanding, changing the regression method only helps to speed up the computation, but not necessarily solve the problem with 99th percentile that p-values for all the factors are 1.0. I wonder how I should interpret the result for 99th percentile, while the results for other percentiles seem to work fine.

Correct me if I?m wrong.

Thank you!

Yunqi
On Nov 16, 2014, at 8:42 AM, Roger <rkoenker at illinois.edu> wrote: