GENERALIZED REGRESSION TREES

被引：54

作者：

CIAMPI, A ^{[1
]}

机构：

[1] MCGILL UNIV,MONTREAL H3Z 2Z3,QUEBEC,CANADA

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 1991年 / 12卷 / 01期

关键词：

DATA ANALYSIS; CLASSIFICATION AND REGRESSION TREES (CART); GENERALIZED LINEAR MODELS (GLIM); TREE-GROWING; PRUNING; CROSS-VALIDATION;

D O I：

10.1016/0167-9473(91)90103-9

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Trees are powerful tools to describe and organize knowledge. As a method of data analysis, trees are used in many disciplines to structure a priori knowledge as well as to formulate and test hypotheses. The author outlines in this paper a general approach to tree-growing whose goal is to predict a parameter of a statistical model, i.e., to develop a general framework for constructing regression trees from data in the context of a generalized regression model (GLIM). Problems of overfit bias and suggestions for further work are discussed.

引用

页码：57 / 78

页数：22

共 20 条

[1]

ALI SM, 1966, J ROY STAT SOC B, V28, P131

[2]

BERAN R, 1977, ANN STAT, V22, P79

[3]

Breiman L, 2017, CLASSIFICATION REGRE, P368, DOI 10.1201/9781315139470

[4] RECPAM - A COMPUTER-PROGRAM FOR RECURSIVE PARTITION AND AMALGAMATION FOR CENSORED SURVIVAL-DATA AND OTHER SITUATIONS FREQUENTLY OCCURRING IN BIOSTATISTICS .1. METHODS AND PROGRAM FEATURES [J].

CIAMPI, A ;

HOGG, SA ;

MCKINNEY, S ;

THIFFAULT, J .

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 1988, 26 (03) :239-256

[5]

Ciampi A, 1987, RECURSIVE PARTITION, P23

[6] SIMULTANEOUS TEST PROCEDURES - SOME THEORY OF MULTIPLE COMPARISONS [J].

GABRIEL, KR .

ANNALS OF MATHEMATICAL STATISTICS, 1969, 40 (01) :224-&

[7]

Gueguen A., 1988, REV STATISTIQUE APPL, V36, P19

[8]

KRZANOWSKI WJ, 1983, BIOMETRIKA, V70, P235

[9] ON INFORMATION AND SUFFICIENCY [J].

KULLBACK, S ;

LEIBLER, RA .

ANNALS OF MATHEMATICAL STATISTICS, 1951, 22 (01) :79-86

[10] EFFICIENT SCREENING OF NONNORMAL REGRESSION-MODELS [J].

LAWLESS, JF ;

SINGHAL, K .

BIOMETRICS, 1978, 34 (02) :318-327

← 1 2 →