PLS fits closer than PCR
- 1 November 1993
- journal article
- research article
- Published by Wiley in Journal of Chemometrics
- Vol. 7 (6) , 551-557
- https://doi.org/10.1002/cem.1180070608
Abstract
PLS1 regression is generally viewed as lying in between PCR and OLS regression. Proof is given that the coefficient of determination,R2, for a PLS multivariate calibration model is at least as high as that for a PCR model with the same number of components. It appears that PLS can be linked to a correlation‐weighted polynomial regression of a constant response on the eigenvalues of the covariance matrix of the predictor variables.Keywords
This publication has 11 references indexed in Scilit:
- A Statistical View of Some Chemometrics Regression ToolsTechnometrics, 1993
- SIMPLS: An alternative approach to partial least squares regressionChemometrics and Intelligent Laboratory Systems, 1993
- The H-principle in modelling with applications to chemometricsChemometrics and Intelligent Laboratory Systems, 1992
- The geometry of 2-block partial least squares regressionCommunications in Statistics - Theory and Methods, 1992
- Calibration modeling by partial least-squares and principal component regression and its optimization using an improved leverage correction for prediction testingChemometrics and Intelligent Laboratory Systems, 1990
- Continuum Regression: Cross-Validated Sequentially Constructed Prediction Embracing Ordinary Least Squares, Partial Least Squares and Principal Components RegressionJournal of the Royal Statistical Society Series B: Statistical Methodology, 1990
- PLS regression methodsJournal of Chemometrics, 1988
- Analysis of two partial-least-squares algorithms for multivariate calibrationChemometrics and Intelligent Laboratory Systems, 1987
- A theoretical foundation for the PLS algorithmJournal of Chemometrics, 1987
- The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized InversesSIAM Journal on Scientific and Statistical Computing, 1984