Cross-validation as the objective function for variable-selection techniques

被引：211

作者：

Baumann, K ^{[1
]}

机构：

[1] Univ Wurzburg, Dept Pharm, D-97074 Wurzburg, Germany

来源：

TRAC-TRENDS IN ANALYTICAL CHEMISTRY | 2003年 / 22卷 / 06期

关键词：

chance correlation; cross-validation; overfitting; permutation test; variable selection;

D O I：

10.1016/S0165-9936(03)00607-1

中图分类号：

O65 [分析化学];

学科分类号：

070302 ; 081704 ;

摘要：

Different methods of cross-validation are studied for their suitability to guide variable-selection algorithms to yield highly predictive models. It is shown that the commonly applied leave-one-out cross-validation has a strong tendency to overfitting, underestimates the true prediction error, and should not be used without further constraints or further validation. Alternatives to leave-one-out cross-validation and other validation methods are presented. (C) 2003 Published by Elsevier Science B.V.

引用

页码：395 / 406

页数：12

共 51 条

[1] Feature selection for structure-activity correlation using binary particle swarms
Agrafiotis, DK
Cedeño, W
[J]. JOURNAL OF MEDICINAL CHEMISTRY, 2002, 45 (05) : 1098 - 1107
[2] RELATIONSHIP BETWEEN VARIABLE SELECTION AND DATA AUGMENTATION AND A METHOD FOR PREDICTION
ALLEN, DM
[J]. TECHNOMETRICS, 1974, 16 (01) : 125 - 127
[3] Use of spectral window preprocessing for selecting near-infrared reflectance wavelengths for determination of the degree of enzymatic retting of intact flax stems
Archibald, DD
Akin, DE
[J]. VIBRATIONAL SPECTROSCOPY, 2000, 23 (02) : 169 - 180
[4] PREDICTIVE ABILITY OF REGRESSION-MODELS .2. SELECTION OF THE BEST PREDICTIVE PLS MODEL
BARONI, M
CLEMENTI, S
CRUCIANI, G
COSTANTINO, G
RIGANELLI, D
OBERRAUCH, E
[J]. JOURNAL OF CHEMOMETRICS, 1992, 6 (06) : 347 - 356
[5] GENERATING OPTIMAL LINEAR PLS ESTIMATIONS (GOLPE) - AN ADVANCED CHEMOMETRIC TOOL FOR HANDLING 3D-QSAR PROBLEMS
BARONI, M
COSTANTINO, G
CRUCIANI, G
RIGANELLI, D
VALIGI, R
CLEMENTI, S
[J]. QUANTITATIVE STRUCTURE-ACTIVITY RELATIONSHIPS, 1993, 12 (01): : 9 - 20
[6] A systematic evaluation of the benefits and hazards of variable selection in latent variable regression. Part II. Practical applications
Baumann, K
von Korff, M
Albert, H
[J]. JOURNAL OF CHEMOMETRICS, 2002, 16 (07) : 351 - 360
[7] Baumann K, 2002, QUANT STRUCT-ACT REL, V21, P507, DOI 10.1002/1521-3838(200211)21:5<507::AID-QSAR507>3.0.CO
[8] 2-L
[9] A systematic evaluation of the benefits and hazards of variable selection in latent variable regression. Part I. Search algorithm, theory and simulations
Baumann, K
Albert, H
von Korff, M
[J]. JOURNAL OF CHEMOMETRICS, 2002, 16 (07) : 339 - 350
[10] An alignment-independent versatile structure descriptor for QSAR and QSPR based on the distribution of molecular features
Baumann, K
[J]. JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 2002, 42 (01): : 26 - 35

← 1 2 3 4 5 6 →