An Enhanced Algorithm for Linear Multivariate Calibration

Analytical Chemistry

Volumn 70, Issue 3, 1998, Pages 623-627

  Berger, Andrew J ^a   Koo, Tae Woong ^a   Itzkan, Irving ^a   Feld, Michael S ^a  

^a MASSACHUSETTS INSTITUTE OF TECHNOLOGY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHM; ARTICLE; CALIBRATION; MULTIVARIATE ANALYSIS; RAMAN SPECTROMETRY; STANDARD;

ALGORITHMS; CALIBRATION; MULTIVARIATE ANALYSIS; SPECTRUM ANALYSIS, RAMAN;

EID: 0031992573     PISSN: 00032700     EISSN: None     Source Type: Journal    
DOI: 10.1021/ac970721p     Document Type: Article

References (8)

1. Haaland, D. M.; Thomas, E. V. Anal. Chem. 1988, 60, 1193-1202.
2. Geladi, P.; Kowalski, B. R Anal. Chim. Acta 1986, 185, 1-17.
3. Thomas, E. V.; Haaland, D. M. Anal. Chem. 1990, 62, 1091-1099.
4. Lorber, A.; Kowalski, B. R. J. Chemom. 1988, 2, 67-79.
5. Sanchez, E.; Kowalski, B. R. J. Chemom. 1988, 2, 247-263.
6. Berger, A. J.; Wang, Y.; Feld, M. S. Appl. Opt. 1996, 35, 209-213.
7. Berger, A. J.; Feld, M. S. Appl. Spectrosc. 1997, 51, 725-732.
8. The procedure is as follows: First, generalize step 1 of the derivation in the Derivation section to remove spectral contributions from multiple species. Principal components of the residual spectra will then describe only the remaining (unknown) background constituents. Using the technique of step 3, orthogonalize the known background spectra to the principal components and to each other. These orthogonalized spectra are then included in the ensemble of spectra V, along with the principal components. In other words, part of V is now constructed from known line shapes, while only the unknown ones are derived from PCA. The derivation then proceeds as before with steps 3 and 4 to calculate the b vector.