Full length paperPartial least squares and compositional data: problems and alternatives
References (40)
- et al.
Chem. Geol.
(1991) Chemom. Intell. Lab. Syst.
(1987)Chemom. Intell. Lab. Syst.
(1988)Chemom. Intell. Lab. Syst.
(1992)Chemom. Intell. Lab. Syst.
(1988)- et al.
Geochem. Cosmochem. Acta
(1987) - et al.
J. Petrol.
(1972) Math. Geol.
(1881)- et al.
J. Sediment. Petrol.
(1968) Math. Geol.
(1976)
Anal. Chem.
Math. Geol.
J. Chemom.
Multivariate Calibration
The Statistical Analysis of Compositional Data
J. Geophys. Res.
Biometrika
Ann. Math. Stat.
Math. Geol.
Cited by (32)
Rapid prediction of essential oils major components by Vis/NIRS models using compositional methods
2022, Results in ChemistryCitation Excerpt :The reasons for this is alr and clr transformations have some drawbacks from their asymmetry and anisometry that make interpretation difficult [11]. Partial least squares (PLS) analysis involving compositional data was first discussed by Hinkle and Rayens [13], althoug it was not done with orthonormal log-ratio coordinates. Egozcue et al. [11] introduced a new family of log-ratio transformations of the simplex, called isometric log-ratio (ILR) because it allows to associate angles and distances in the simplex, following the Aitchison geometry, to angles and distances in real space.
Chemometric evaluation of alfalfa sprouting impact on its metabolic profile using HPTLC fingerprint-efficacy relationship analysis modelled with partial least squares regression
2020, Journal of Pharmaceutical and Biomedical AnalysisCitation Excerpt :So that, a set of unsupervised and supervised chemometric tools was performed [21]. The corner stones of the unsupervised pattern recognition methods are the Principal Component Analysis (PCA) and the Hierarchical Clustering Analysis (HCA) which are the exploratory algorithms that require little or no prior knowledge of the data set [22], Partial least square regression (PLS) in case of linear supervised method [23]. In the present study, the impact of sprouting of alfalfa seeds along eleven days on its holistic metabolic pole was tracked with the aim of discovery of metabolically- significant markers.
Olive oil nutritional labeling by using Vis/NIR spectroscopy and compositional statistical methods
2019, Innovative Food Science and Emerging TechnologiesCitation Excerpt :For a case in which the composition played the role of explanatory variable, Palarea-Albaladejo, Rooke, Nevison, and Dewhurst (2017) developed a compositional mixed model to explain methane production from ruminal volatile fatty acids in cattle, along with other diet and animal covariates. Partial least squares (PLS) analysis involving compositional data was first discussed in chemometrics by Hinkle and Rayens (1995), although it was not done in terms of orthonormal log-ratio coordinates since this was a later development introduced by Egozcue, Pawlowsky-Glahn, Mateu-Figueras, and Barceló-Vidal (2003). An application of PLS modelling to discriminant analysis (PLS-DA), which treated the metabolomics profiles as compositions via log-ratios, can be found in Kalivodová et al. (2015).
The Basis Function Regression in pharmaceutical analysis. Theory and example of application
2008, Journal of Pharmaceutical and Biomedical AnalysisCitation Excerpt :The Principal Component Regression (PCR) [6] uses the k first principal components of the calibration matrix as the regressors. The Partial Least Squares (PLS) [7] takes into account also the response and maximises the covariance between them. The Cyclic Subspace Regression (CSR) [8] and Continuum Power Regression (CPR) [9] are the generalized techniques.
Reciprocal curves
2006, Computational Statistics and Data AnalysisPartial least squares as a target-directed structure-seeking technique
2004, Chemometrics and Intelligent Laboratory Systems