Abstract
We have begun introducing complex-valued principal component regression (PCR) into spectroscopy. Unlike traditional methods that are constrained to either the real or imaginary axis, this approach allows principal components (PCs) to span the entire complex plane. While this added flexibility enhances modeling capabilities, it also introduces challenges, as existing tools often fail to identify optimal solutions. To address this, we explored two different strategies for computing eigenvectors. The most natural approach is to apply singular value decomposition (SVD) directly to the matrix of complex refractive index spectra. As an alternative, we combined the eigenvectors of the imaginary parts determined by SVD with their Kramers–Kronig transforms, which resulted in 2 N possible superpositions for N PCs. Although the optimal solution may still be unknown, the proposed second method for complex-valued PCR consistently outperformed conventional PCR in the systems investigated. This highlights its potential to enhance data analysis in infrared and Raman spectroscopy.
This is a visual representation of the abstract.
Keywords
Get full access to this article
View all access options for this article.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
