REFERENCE VALUES OBTAINED BY KERNEL-BASED ESTIMATION OF QUANTILE REGRESSIONS

Citation
Gr. Ducharme et al., REFERENCE VALUES OBTAINED BY KERNEL-BASED ESTIMATION OF QUANTILE REGRESSIONS, Biometrics, 51(3), 1995, pp. 1105-1116
Citations number
34
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
ISSN journal
0006341X
Volume
51
Issue
3
Year of publication
1995
Pages
1105 - 1116
Database
ISI
SICI code
0006-341X(1995)51:3<1105:RVOBKE>2.0.ZU;2-3
Abstract
In this paper, we study the problem of estimating non-parametrically a quantile regression curve as it applies to computing reference values . We propose an automatic procedure that uses a symmetrized nearest-ne ighbor kernel estimator of conditional distributions. We also discuss ways of measuring the dispersion of the quantile regression estimator. One is based on the asymptotic distribution of such quantiles, while the other relies on the bootstrap method. The results of a small simul ation study show that the methods of the paper perform rather well eve n in a situation where a good parametric solution is available. As an example, we analyze a small part of a data set that was collected to e stablish reference values for blood velocity in different parts of the umbilical cord of human fetuse as they grow toward birth.