INCLUSION OF THE STANDARD-DEVIATION OF DATA IN PRINCIPAL COMPONENT ANALYSIS - A GRAPHICAL APPROXIMATION

The adsorption capacity and specific adsorption surface area of 13 ant i-hypoxia drugs were determined in three chromatographic systems using methanol-carbon tetrachloride, chloroform-carbon tetrachloride and ac etonitrile-carbon tetrachloride mixtures as eluents. The retention beh aviours of the anti-hypoxia drugs were compared using principal compon ent analysis (PCA). A graphical approximation was used for the inclusi on of the standard deviations of both the variables and observations i n PCA and the results were visualized by two-dimensional nonlinear map ping and cluster analysis. The results indicated that the graphical ap proximation can be successfully used for the inclusion of the standard deviation of data in PCA calculations. Nonlinear mapping and cluster analysis resulted in similar, but not identical, classification of dru gs and chromatographic systems, indicating that each multivariate meth od can be successfully used for the comparison of solutes and chromato graphic systems.