An estimator of a conditional quantile in the presence of auxiliary information |
| |
| Institution: | 1. Center for Interdisciplinary Research in Biology (CIRB), College de France - CNRS, INSERM, Université PSL, 11 Place Marcelin Berthelot, Paris 75005, France;2. Institut de Génomique Fonctionnelle de Lyon (IGFL), ENS de Lyon, CNRS, Univ Lyon 1, 32-34 Avenue Tony Garnier, Lyon 69007, France;3. R&D Department, SILAB, ZI de la Nau, Saint Viance 19240, France;4. Sorbonne Université, Collège Doctoral, 15 rue de l''Ecole de Médecine, Paris 75006, France |
| |
| Abstract: | In this paper, a new estimator for a conditional quantile is proposed by using the empirical likelihood method and local linear fitting when some auxiliary information is available. The asymptotic normality of the estimator at both boundary and interior points is established. It is shown that the asymptotic variance of the proposed estimator is smaller than those of the usual kernel estimators at interior points, and that the proposed estimator has the desired sampling properties at both boundary and interior points. Therefore, no boundary modifications are required in our estimation. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
