PLS path modelling: computation of latent variables with the estimation mode B

被引:51
作者
Hanafi, Mohamed [1 ]
机构
[1] INRA, ENITIAA, Unite Sensometrie & Chimiometrie, F-44322 Nantes, France
关键词
path modelling; latent variables; structural modelling; covariance structure; monotonically convergence;
D O I
10.1007/s00180-007-0042-3
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
PLS Path modelling has several interesting advantages compared to other existing approaches traditionally used for structural modelling. However, the lack of convergence properties of the existing iterative procedures for the computation of the latent variables, has always been considered as a major drawback. The convergence is stated only in practice. The present paper shows that when the estimation mode B is chosen for all blocks, the iterative procedure for the computation of latent variables proposed by Wold (in Encyclopaedia of statistical sciences, vol 6. Wiley, New York, pp. 581-591, 1985) is monotonically convergent.
引用
收藏
页码:275 / 292
页数:18
相关论文
共 10 条
[1]  
BOOKSTEIN FL, 1982, SYSTEMS INDIRECT OBS, V2, P55
[2]  
Chin W., 2001, PLS-Graph User's Guide: CT Bauer College of Business
[3]   A GENERAL METHOD FOR ANALYSIS OF COVARIANCE STRUCTURES [J].
JORESKOG, KG .
BIOMETRIKA, 1970, 57 (02) :239-&
[4]  
Lohmoller J.B., 1989, LATENT VARIABLES PAT
[5]  
LOHMOLLER JB, 1984, LVPLS PROGRAMM MANUE
[6]  
Mathes H, 1993, STAT MODELLING LATEN
[7]  
Ringle C.M., 2005, SMARTPLS
[8]   PLS path modeling [J].
Tenenhaus, M ;
Vinzi, VE ;
Chatelin, YM ;
Lauro, C .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2005, 48 (01) :159-205
[9]  
Wold H., 1985, Encyclopedia of Statistical Sciences, P581, DOI [10.1002/0471667196.ESS1914.PUB2, DOI 10.1002/0471667196.ESS1914.PUB2]
[10]  
Wold H., 1982, SYSTEMS INDIRECT OBS, P1