Sliced inverse regression in reference curves estimation

被引：16

作者：

Gannoun, A

Girard, S

Guinot, C

Saracco, J ^{[1
]}

机构：

[1] Univ Montpellier 2, Lab Probabil & Stat, Pl Eugene Bataillon, F-34095 Montpellier 5, France

[2] Howard Univ, Natl Human Genome Ctr, Stat Genet & Bioinformat Unit, Washington, DC 20059 USA

[3] Univ Grenoble 1, SMSLMC, F-38041 Grenoble 9, France

[4] CERIES, F-92521 Neuilly Sur Seine, France

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 2004年 / 46卷 / 01期

关键词：

conditional quantiles; dimension reduction; kernel estimation; semiparametric method; reference curves; sliced inverse regression (SIR);

D O I：

10.1016/S0167-9473(03)00141-5

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

In order to obtain reference curves for data sets when the covariate is multidimensional, a new procedure is, proposed. This procedure is based on dimension-reduction and non-parametric estimation of conditional quintiles. This semiparametric approach combines sliced inverse regression (SIR) and a kernel estimation of conditional quintiles. The asymptotic convergence of the derived estimator is shown. By a simulation study, this procedure is compared to the classical kernel icon-parametric one for different dimensions of the covariate. The semiparametric estimator shows the best performance. The usefulness of this estimation procedure is illustrated on a real data set collected in order to establish reference curves for biophysical properties of the skin of healthy French women. (C) 2003 Elsevier B.V. All rights reserved.

引用

页码：103 / 122

页数：20

共 54 条

[1]

Aragon Y, 1997, COMPUTATION STAT, V12, P109

[2] Asymptotic normality of convergent estimates of conditional quantiles [J].