Usefulness of Multi-Criteria Taxonomy in Corporate Bankruptcy Prediction - The Polish Experience

One of the essential elements of company's economic and financial evaluation is ratio analysis. It includes computation and interpretation of multiple bankruptcy risk ratios. When applying accounting ratios in a credit risk evaluation analysts typically either compute and interpret several individual ratios (a univariate approach) or apply some pre-estimated parametric multivariable econometric tools, such as the Altman model or discriminant function. The problem with the first approach is that it does not take into account the inter-relationships between the individual ratios. The latter approach, in turn, calls for statistical models, whose parameters may not be stable in time and in space. In this paper we empirically examine the usefulness of the alternative approach, that is simple taxonomic ranking, in corporate bankruptcy risk quantification (on the basis of data from the Polish stock market). We found that this simple non-parametric multivariate method outperforms all other tested approaches (including logit model) in discriminating between bankrupt and healthy firms