COMPARISON OF CHEMOMETRIC MODELS FOR QSAR

被引：32

作者：

CRUCIANI, G

BARONI, M

BONELLI, D

CLEMENTI, S

EBERT, C

SKAGERBERG, B

机构：

[1] UNIV PERUGIA,DEPARTIMENTO CHIM,I-06100 PERUGIA,ITALY

[2] UNIV TRIESTE,IST CHIM FARMACEUT,I-34128 TRIESTE,ITALY

[3] UMEA UNIV,INST CHEM,CHEMOMETR RES GRP,S-90187 UMEA,SWEDEN

来源：

QUANTITATIVE STRUCTURE-ACTIVITY RELATIONSHIPS | 1990年 / 9卷 / 02期

关键词：

chemometrics; multiple linear regression; overfitting; PLS regression; prediction errors; QSAR models;

D O I：

10.1002/qsar.19900090205

中图分类号：

R914 [药物化学];

学科分类号：

100701 ;

摘要：

Chemometric methods suitable for QSAR modelling, namely partial least squares (PLS) and multiple regression analysis (MRA), are compared in terms of their prediction abilities by means of the appropriate statistical parameter called standard deviation of errors of predictions (SDEP). The results show the reliability of the PLS approach and point out the risk of overfitting with MRA, which may give rise to unreliable predictions. Copyright © 1990 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim

引用

页码：101 / 107

页数：7

共 27 条

[1]

BERTUCCIOLI M, 1984, VINI ITALIA, V26, P31

[2]

BONELLI D, IN PRESS IL FARMACO

[3]

CHARTON M, 1985, QUANT STRUCT-ACT REL, V4, P29

[4] SOME APPLICATIONS OF THE PARTIAL LEAST-SQUARES METHOD [J].