Peculiar behavior of Japanese bankrupt firms: discovered by AI-based data mining technique

The study presents some empirical results of a study regarding financial ratios as predictors of Japanese corporate bankruptcy based on a large sample of bankrupt and non-bankrupt firms' financial data. In this study, variables as predictors of bankruptcy were selected by two data mining techniques: CART and Stepwise. After the selection of a set of variables for the two techniques, the discriminant power of each set was compared to identify the best set of financial ratios to predict corporate bankruptcy. However, the importance of this kind of study is not to develop the prediction model with selected variables, but to discover the turning point of whether a firm is going into bankruptcy or not. The study introduces how to find out the premonitory symptoms of Japanese corporate bankruptcy from selected financial variables.