论文信息 - Improved selectivity in spectroscopy by multivariate calibration

Improved selectivity in spectroscopy by multivariate calibration

This paper illustrates some advantages of indirect multivariate calibration over conventional calibration: selectivity enhancement and outlier detection. Partial least squares (PLS) calibration is applied for quantitative analysis in the presence of interferences that would make both conventional single‐wavelength calibration and direct multicomponent analysis impossible. The PLS algorithm is illustrated graphically, and the importance of outlier detection is demonstrated.

Tormod Næs | Harald Martens | Terje V. Karstang | T. Næs | H. Martens

[1] S. Wold,et al. The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized Inverses , 1984 .

[2] P. Williams,et al. Near-Infrared Technology in the Agricultural and Food Industries , 1987 .

[3] E. B. Wilson. Mathematics and Statistics , 1930 .

[4] H. Martens,et al. Comparison of Linear Statistical Methods for Calibration of Nir Instruments , 1986 .

[5] B. Kowalski,et al. Partial least-squares regression: a tutorial , 1986 .

[6] Tormod Næs,et al. Multivariate Calibration When the Error Covariance Matrix Is Structured , 1985 .

[7] Tormod Næs,et al. Multivariate calibration. II. Chemometric methods , 1984 .

[8] Tormod Næs,et al. Comparison of prediction methods for multicollinear data , 1985 .