FITTING ADDITIVE-MODELS TO REGRESSION DATA - DIAGNOSTICS AND ALTERNATIVE VIEWS

被引：30

作者：

BREIMAN, L ^{[1
]}

机构：

[1] UNIV CALIF BERKELEY,BERKELEY,CA 94720

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 1993年 / 15卷 / 01期

基金：

美国国家科学基金会;

关键词：

ADDITIVE MODELS; SPLINES; CROSS-VALIDATION; KNOT DELETION; STABILITY;

D O I：

10.1016/0167-9473(93)90217-H

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

A method is given for fitting additive models to multivariate regression type data. The emphasis is on diagnostics analogous to the concept of influence and on alternative models. These are illustrated in three case studies. Simulations show excellent predictive accuracy as compared with existing methods.

引用

页码：13 / 46

页数：34

共 23 条

[11]

FRIEDMAN JH, 1982, ORION003 STANF U STA

[12] BOOTSTRAPPING IN NONPARAMETRIC REGRESSION - LOCAL ADAPTIVE SMOOTHING AND CONFIDENCE BANDS [J].