A NOTE ON ASYMPTOTICS FOR QUANTILE SMOOTHING SPLINES

被引：0
|
作者：
BOSCH, RJ [1 ]
机构：
[1] HARVARD UNIV,SCH PUBL HLTH,DEPT BIOSTAT,BOSTON,MA 02115
来源：
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 1994年 / 23卷 / 11期
基金：
美国国家卫生研究院;
关键词：
SMOOTHING SPLINES; NONPARAMETRIC REGRESSION; QUANTILE REGRESSION;
D O I：
10.1080/03610929408831434
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
When cubic smoothing splines are used to estimate the conditional quantile function, thereby balancing fidelity to the data with a smoothness requirement, the resulting curve is the solution to a quadratic program. Using this quadratic characterization and through comparison with the sample conditional quantiles, we show strong consistency and asymptotic normality for the quantile smoothing spline.
引用
收藏
页码：3075 / 3083
页数：9
相关论文
共 50 条
12345