DISTRIBUTED COMPUTING IN MULTI-GROUP STRUCTURAL EQUATION MODEL
This paper firstly proposes an improved partial least square(PLS) algorithm in structural equation model (SEM) using a suitable iterative initial value with constraint of unit vector.The algorithm enhances the convergence rate greatly and its convergency is illuminated. Then multi-group SEM is investigated and distributed computing is adopted to calculate all the coefficients in SEM of each group, which results in considerable timesavings when compared with traditional approach. Furthermore, a uniform model is built using the generalized linear model with convex constraint and an algorithm for the multi-group SEM is presented. The results of this paper have been received in software DASC.
SEM PLS Multi-Group Model Convergency Distributed Computing
Hengqing Tong Ziqing Li QiaoLing Tong
Department of Mathematics, Wuhan University of Technology Wuhan, Hubei Province 430070, P.R.China Department of Electronic Science and Technology, Huazhong University of Science and Technology Wuhan
国际会议
杭州
英文
144-148
2006-10-12（万方平台首次上网日期，不代表论文的发表时间）