Predicting financial distress of agriculture companies in EU

Abstract

The objective of this paper is prediction of financial distress (default of payment or insolvency) of 250 agriculture business companies in EU from which 62 companies defaulted in 2014 with respect to lag of the used attributes. From many types of classification models we chose Logistic regression, Support vector machines method with RBF ANOVA kernel, Decision trees and Adaptive boosting based on decision trees to acquire the best results. From the results it is obvious that with the rising distance to the bankruptcy there drops average accuracy of financial distress prediction and there is a greater difference between active and distressed companies in terms of liquidity, rentability and debt ratios. The Decision trees and Adaptive boosting offer better accuracy for distress prediction than SVM and logit methods, what is comparable to previous studies. From overall of 15 accounting variables, we construct classification trees by Decision trees with inner feature selection method for better vizualization, what reduce full data set only to 1 or 2 attributes: ROA and Long-term debt to Total assets ratio in 2011, ROA and Current ratio in 2012, ROA in 2013 for discrimination of distressed companies.