Segmentation of Firms by Means of Classification Trees

The objective of the paper was to present the utility and applicability of the method of generating classification trees for the purposes of segmentation of firms by their economic standing, i.e. their financial and assets condition. The method of classification tree generation belongs to the group of the “data mining” methods that permit to find out, basing on large data sets, the relationships and links among data. Variables used to classify the firms were financial and assets indices, indices of the firms’ position in the capital market, as well as values from financial statements (balance sheet, profit and loss account, cash flow account). Different algorithms were used to generate the classification trees: Chaid, Exhaustion Chaid, C&RT, Quest. In the paper, examples were presented of applying the classification trees to segmentation of firms. Advantages and drawbacks of perforrning the segmentation of firms by means of the classification trees were discussed.