Nonlinear multivariate curve resolution alternating least squares (NL-MCR-ALS)†

摘要: Bilinearity is the basic principle of multivariate curve resolution. In this paper, we consider a case when premise violated. We demonstrate that alternating least squares approach can still be used to solve problem. The developed theory applied calibration spectral data includes so-called saturated peaks, which are flattened because samples with ultrahigh absorbance. in spite serious violations Lambert–Beer law, results prediction quite satisfactory, and accuracy better than other competing methods. Copyright © 2014 John Wiley & Sons, Ltd.