| We consider nonparametric estimation of conditional medians when both explanatory variables and error terms exhibit long-range dependence. We do not assume the independence between explanatory variables and error terms. The estimator is based on minimizing the locally weighted sum of absolute deviations. We present the asymptotic distribution of the local linear estimator. The asymptotic distribution depends on the strength of long-range dependence of the underlying bivariate linear process when the order of the bandwidths is fixed as h=cn^{-1/5}, where n is the sample size. The paper is available at http://www.econ.hit-u.ac.jp/~kenkyu/jpn/pub/dp-top.htm |