USE OF EIGENVALUES FOR DETERMINING THE NUMBER OF COMPONENTS IN WINDOWFACTOR-ANALYSIS OF SPECTROSCOPIC AND CHROMATOGRAPHIC DATA

Three methods for determining eigenvalues, namely (I) mean-centred seq uential scores, (2) mean-centred sequential loadings and (3) uncentred data, are described, and compared on simulations and real data of two -way data matrices. The sequential direction may be time (high-perform ance liquid chromatography (HPLC)) or wavelength (infrared spectroscop y (IR)), with the non-sequential direction wavelength (HPLC) or sample number (IR). It is shown that method 1 is very dependent on peak shap es. Method 2 works well except when there is low Variability in the no n-sequential direction. Method 3 performs poorly in the presence of he teroscedastic noise. It is concluded that standard statistical approac hes for eigenanalysis must be used with caution in chemometric methods for window factor analysis (WFA).