Improved inference in nonparametric regression using L(k)-smoothing splines

被引:20
作者
Abramovich, F [1 ]
Steinberg, DM [1 ]
机构
[1] TEL AVIV UNIV,RAYMOND & BEVERLY SACKLER FAC EXACT SCI,DEPT STAT & OPERAT RES,IL-69978 RAMAT AVIV,ISRAEL
关键词
Bayesian linear model; confidence interval; L-spline; variable smoothing parameter;
D O I
10.1016/0378-3758(95)00021-6
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
Smoothing splines are one of the most popular approaches to nonparametric regression. Wahba (J. Roy. Statist. Sec. Ser. B 40 (1978) 364-372; 45 (1983) 133-150) showed that smoothing splines are also Bayes estimates and used the corresponding prior model to derive interval estimates for the regression function. Although the interval estimates work well on a global basis, they can have poor local properties. The source of this problem is the use of a global smoothing parameter. We introduce the notion of L(k)- smoothing splines. These splines allow for a variable smoothing parameter and can substantially improve local inference.
引用
收藏
页码:327 / 341
页数:15
相关论文
共 27 条
[1]  
ANSLEY CF, 1993, BIOMETRIKA, V80, P75, DOI 10.1093/biomet/80.1.75
[2]   BAYESIAN APPROACH TO MODEL INADEQUACY FOR POLYNOMIAL REGRESSION [J].
BLIGHT, BJN ;
OTT, L .
BIOMETRIKA, 1975, 62 (01) :79-88
[3]  
BUJA A, 1989, ANN STAT, V17, P453, DOI 10.1214/aos/1176347115
[4]  
CARTER CK, 1992, J ROY STAT SOC B MET, V54, P773
[5]   SMOOTHING NOISY DATA WITH SPLINE FUNCTIONS [J].
WAHBA, G .
NUMERISCHE MATHEMATIK, 1975, 24 (05) :383-393
[6]  
Eppright E S, 1972, World Rev Nutr Diet, V14, P269
[7]  
Eubank R.L., 1988, SPLINE SMOOTHING NON
[8]   FITTING SEGMENTED POLYNOMIAL REGRESSION MODELS WHOSE JOIN POINTS HAVE TO BE ESTIMATED [J].
GALLANT, AR ;
FULLER, WA .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1973, 68 (341) :144-147
[9]  
Green Peter J., 1993, NONPARAMETRIC REGRES
[10]   SMOOTHING NOISY DATA WITH SPLINE FUNCTIONS [J].
HUTCHINSON, MF ;
de Hoog, FR .
NUMERISCHE MATHEMATIK, 1985, 47 (01) :99-106