Regularised discriminant analysis (RDA) - modelling for the binary discrimination between pollution types

Discrimination between different types of pollution is carried out with RDA. A description of the most important characteristics of the model is presented. Attention is paid to the four extreme discrimination models that are special cases of RDA and which are the well known linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA), the nearestmeans classifier (NM) and the weighted nearest means-classifier (WNM). A comparison among RDA and partial least squares (PLS) is attempted and some first considerations are proposed. Feature selection to determine the most important variables for the discrimination is studied.