Elastic Component Regression (ECR) for uncovering the path from PCR to PLS 1.0

ECR is a new method for regression analysis, which employs a supervising alpha to supervise the X-matrix decomposition. When alpha=0, ECR coincides with principal component regression (PCR), when alpha=1, ECR coincides with partial least squares (PLS), when alpha is between 0 and 1, ECR could identify an infinite number of transitional models between PCR and PLS which collectively indicate a path that connects PCR and PLS.For details, please refer to:Hong-Dong Li, Yi-Zeng Liang, Qing-Song Xu, Uncover the path from PCR to PLS via elastic component regression, Chemometrics and Intelligent Laboratory Systems, 2010, 104(2)341-346

Requirements:	No special requirements
Platforms:	Matlab
Keyword:	Collectively,  Connects,  Details,  Hongdong,  Models,  Number,  Refer,  Transitional
Users rating:	0/10