Latent Variable Path Modeling with Partial Least Squares
Partial Least Squares (PLS) is an estimation method and an algorithm for latent variable path (LVP) models. PLS is a component technique and estimates the latent variables as weighted aggregates. The implications of this choice are considered and compared
- PDF / 27,839,085 Bytes
- 284 Pages / 482.151 x 691.937 pts Page_size
- 93 Downloads / 437 Views
Latent Variable Path Modeling with Partial Least Squares
Springer-Verlag Berlin Heidelberg GmbH
Dr. Jan-Bernd Lohmoller Free University Berlin Dept. of Political Science (Otto-Suhr-Institute) Ihnestra13e 21 D-I000 Berlin 33, FRG
Min Ollern toudacht
ISBN 978-3-642-52514-8
CIP-Titelaufnahme der Deutschen Bibliothek Lohmăller, Jan-Bernd: Latent variable path model ing with partial least squares / JanBernd Lohmăller. ISBN 978-3-642-52514-8 ISBN 978-3-642-52512-4 (eBook) DOI 10.1007/978-3-642-52512-4
This work is subject ta copyright. AII rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translatioI), reprinting, reuse of illustrations, re citation, broadcasting, reproduction on microfilms Of in other ways, and starage in data banks. Duplication of this publication or parts thereof is on1y permitted under the provisions of the German Copyright Law of September 9, 1965, in its version of June 24,1985, and a copyright fee must always be paid. Violations fali under the prosecution act of the German Copyright Law. © Springer-Verlag Berlin Heidelberg 1989 Originally published by Physica-Verlag Heidelberg in 1989 The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regu1ations and therefore free for general use. TEX typeset by J. B. L.
712017130-543210
Preface I am greatly indepted to my teachers - and collegues - Professors Heinz Mandl, Rolf Oerter, Rainer B. Pelka, and Jiirgen W. Falter, who gave me unceasing encouragement in my work through their useful! questioning and their confronting me with many problems of a challenging nature over a period of many years. Since 1978 when I first heard a lecture by Professor Herman Wold, I have been excited and captivated by the potentialities preferred by his PLS modeling method. In the ensuing years of collaboration I not only was introduced into and motivated by his scholarly life-style but also personally moved by his attentive and generous attitude. Thanks to his constant inspiration, criticism and encouragement and his "Advice to a Young Scientist" he became what I am proud to call my Doktorvater. Experience with a particular statistical method derives from various sources, namely (a) mathematical reasoning, (b) programming, (c) application and (d) teaching. All four types of experience leave their marks on the language, style, presentation, and type of reasoning in this dissertation: (a) An attempt has been made to make the mathematical reasoning the core of this monograph. (b) The program PLS 1.8 (Lohmoller 1981a) documents how I understand PLS and is understandable to every computer but not, unfortunately, to every reader. So it has to be explained avoiding computer jargon. (c) With every new application new insight is obtained as to what the method is able to do and what not, which may, or may not, require changes in the program and the mathematical treatment. (d) Teaching pr
