Variable Selection in PLSR Based on Gram-Schmidt Process

摘要：

Partial Least-Squares (PLS) is favourable in linear regression modelling under the condition of multicollinearity. Unfortunately, it indicates that the multicollinearity in independent variable sets in PLS can also obviously affect the deriving of components and the regression parameters. In this paper, the Gram-Schmidt process is adopted in die component extracting process of PLS modelling. By applying such modelling method, PLSR can not only preserve the information of independent sets to the most degree, but also exclude the multicollinearity among the variables.

关键词： partial least-squares regression Gram-Schmidt process variable selection dimension reduction

作者: Chen Meiling

作者单位: Beijing University of Aeronautics and Astronautics, Beijing 100191, China

会议类型: 国际会议

会议名称: The 6th International Conference on Partial Least Squares and Related Methods(第六届偏最小二乘及相关方法国际会议)

会议地点: 北京

会议语种:英文

页码: 49-53

在线出版日期: 2009-09-04（万方平台首次上网日期，不代表论文的发表时间）