Monotone shrinkage of trees

被引：15

作者：

LeBlanc, M

Tibshirani, R

机构：

[1] Fred Hutchinson Canc Res Ctr, Seattle, WA 98109 USA

[2] Univ Toronto, Dept Publ Hlth Sci, Toronto, ON M5S 1A8, Canada

[3] Univ Toronto, Dept Stat, Toronto, ON M5S 1A8, Canada

来源：

JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS | 1998年 / 7卷 / 04期

关键词：

CART; classification and regression trees; LASSO;

D O I：

10.2307/1390674

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：

We investigate a new method for regression trees which obtains estimates and predictions subject to constraints on the coefficients representing the effects of splits in the tree. The procedure leads to both shrinking of the node estimates and pruning of branches in the tree and for some problems gives better predictions than cost-complexity pruning used in the classification and regression tree (CART) algorithm. The new method is based on the least absolute shrinkage and selection operator (LASSO) method developed by Tibshirani.

引用

页码：417 / 433

页数：17

共 5 条

[1]

Becker R. A., 1988, NEW S LANGUAGE

[2]

Friedman JH., 1984, BIOMETRICS, V40, P874, DOI [DOI 10.2307/2530946, 10.2307/2530946]

[3] HEDONIC HOUSING PRICES AND DEMAND FOR CLEAN-AIR [J].

HARRISON, D ;

RUBINFELD, DL .

JOURNAL OF ENVIRONMENTAL ECONOMICS AND MANAGEMENT, 1978, 5 (01) :81-102

[4]

HASTIE TJ, 1991, SHRINKING TREES

[5] Regression shrinkage and selection via the Lasso [J].

Tibshirani, R .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 1996, 58 (01) :267-288

← 1 →