Selekcja akcji z wykorzystaniem modeli dyskryminacyjnych oraz optymalizacji podziału zbioru obiektów

This article consists primarily of discriminant analysis, which could be used as a tool to guide the investors to choose appropriate shares. Author used data on the financial situation of various companies from three various branches. The objects in the examination were descibed through set of financial indicators. The purpose of the discriminative models was to select the companies that had a higher return rate than average in a three month period after publishing quarterly financial reports. The use of the k-means method amends quality of the discriminant model improving the chance to figure out an effective prognostic tool.