MULTIVARIATE-ANALYSIS OF AQUATIC TOXICITY DATA WITH PLS

A common task in data analysis is to model the relationships between t wo sets of variables, the descriptor matrix X and the response matrix Y. A typical example in aquatic science concerns the relationships bet ween the chemical composition of a number of samples (X) and their tox icity to a number of different aquatic species (Y). This modelling is done in order to understand the variation of Y in terms of the variati on of X, but also to lay the ground for predicting Y of unknown observ ations based on their known X-data. Correlations of this type are usua lly expressed as regression models, and are rather common in aquatic s cience. Often, however, the multivariate X and Y matrices invalidate t he use of multiple linear regression (MLR) and call for methods which are better suited for collinear data. In this context, multivariate pr ojection methods represent a highly useful alternative, in particular, partial least squares projections to latent structures (PLS). This pa per introduces PLS, highlights its strengths and presents applications of PLS to modelling aquatic toxicity data. A general discussion of re gression, comparing MLR and PLS, is provided.