An interior point algorithm for nonlinear quantile regression

A new algorithm for computing quantile regression estimates for problems in which the response function is nonlinear in parameters is described. The nonlinear l1 estimation problem is a special (median) case. The algorithm is closely related to recent developments on interior point methods for solving linear programs. Performance of the algorithm on a variety of test problems including the censored linear quantile regression problem of Powell (1986) is reported.