Financial distress prediction across firms

Abstract

One of the most important events in a firm’s life is financial distress, which can propel sectors into financial and sustainable growth problems. Moreover, independent variables in the background of financial distress are accounting ratios, which are extracted from financial statements and macroeconomic variables that are mostly beyond the control of a firm or sector. The current study analysed the information related to a sample of 300 public Iranian companies, during the periods of 2000-2007 and 2009-2016. Logistic regression and decision trees were applied to the prediction of financial distress. It was found that the profitability, liquidity, leverage, interest rate, cash flow, accruals, and GDP were statistically significant in distinguishing distressed from non-distressed firms across sectors. The obtained results showed that the predictive performance of a DT model was more successful than the other model.