Iterative deflation algorithm, eigenvalue equations, and PLS2. (27th August 2019)

• Record Type: Journal Article
• Authors: Stocchero, Matteo
• Is Part Of: Journal of chemometrics. Volume 33:Number 10(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-08-27
• Subjects: post‐transformation of PLS2 -- principal component regression -- projection to latent structures regression; Chemistry -- Mathematics -- Periodicals; Chemistry -- Statistical methods -- Periodicals; 542.85
• DOI: 10.1002/cem.3144
• Languages: English
• ISSNs: 0886-9383

Abstract

Abstract: PLS2 is probably the most used algorithm to perform projection to latent structures regression in the case of multivariate response. However, several criticisms pointed to the theoretical limits of its original formulation, highlighting the need of a more robust foundation within the theory of regression analysis. The iterative deflation algorithm is here introduced as a starting point to obtain a family of regression methods, which includes PLS2, principal component regression (PCR), and elastic component regression (ECR), where different eigenvalue equations are used to calculate the weight vectors. Within this framework, an original portrait of PLS2 is drawn. The main mathematical properties useful to understand what PLS2 is and how PLS2 behaves are derived. A new regression method called iterative deflation algorithm‐based regression (IDAR) is introduced to describe the limit behaviour of PLS2, PCR, and ECR. The post‐transformation method is presented as a general property of the iterative deflation algorithm. Two data sets, one simulated and the other experimental, are investigated to illustrate the main properties of PLS2. Abstract : An original portrait of PLS2 regression based on the iterative deflation algorithm and a well‐defined eigenvalue equation is drawn. The main properties of PLS2 are derived, and the post‐transformation method is presented as a general property of the iterative deflation algorithm. Iterative deflation algorithm‐based regression (IDAR) is introduced as limit case of PLS2, PCR, and ECR when the number of latent variable is the maximum.