A couple of weeks back Roger Koenker reminded me about my post comparing quantreg vs. Rmosek. In Angrist’s application, Rmosek was faster than the traditional Barrondale and Roberts. However, Roger was curious about how it would compare to the interior point algorithms implemented using FORTRAN in <a href=“https://cran.r-project.org/web/packages/quantreg/quantreg.pdf" …