The Relevance of Trend Variables for the Prediction of Corporate Crises and Insolvencies

This study investigated the potential of a specific trend, defined as the relative change of accounting ratios for two consecutive years, to improve the classification accuracy and model performance of insolvency prediction models based on multivariate linear discriminant analysis. The results show that the respective trend can include information from both consecutive years, but this informational content could not be exploited to improve early detection of corporate crises and insolvencies.