Variable influence on projection (VIP) for orthogonal projections to latent structures (OPLS)

• Published in: Journal of chemometrics Vol. 28; no. 8; pp. 623 - 632
• Main Authors: Galindo-Prieto, Beatriz, Eriksson, Lennart, Trygg, Johan
• Format: Journal Article
• Language: English
• Published: Chichester Blackwell Publishing Ltd 01-08-2014; Wiley Subscription Services, Inc
• Subjects: Analytical Chemistry; analytisk kemi; Chemometrics; Comparative analysis; Computer Science; datalogi; Forecasting; Formalism; Mathematical models; OPLS; PLS; Projection; Statistics; statistik; variable influence on projection; variable selection; Variables; VIP
• ISSN: 0886-9383; 1099-128X; 1099-128X
• DOI: 10.1002/cem.2627

A new approach for variable influence on projection (VIP) is described, which takes full advantage of the orthogonal projections to latent structures (OPLS) model formalism for enhanced model interpretability. This means that it will include not only the predictive components in OPLS but also the orthogonal components. Four variants of variable influence on projection (VIP) adapted to OPLS have been developed, tested and compared using three different data sets, one synthetic with known properties and two real‐world cases. Copyright © 2014 John Wiley & Sons, Ltd.
A new approach for variable influence on projection is described, which takes full advantage of the orthogonal projections to latent structures (OPLS) model formalism for enhanced model interpretability. This means that it will include not only the predictive components in OPLS but also the orthogonal components. Four variants of variable influence on projection adapted to OPLS have been developed, tested, and compared using three different data sets, one synthetic with known properties and two real‐world cases.