Designing a model for predicting financial distress in Iran's business ecosystem using deep learning network

Financial distress of companies leads to waste of resources and failure to take advantage of investment opportunities. When a company is in financial trouble, it is expected to face one of these two possible conflicts, a lack of cash on the assets side of the balance sheet or an inflation of liabilities on the left side of the balance sheet. Most of the Iranian companies prefer to convert their cash into other assets due to the existing inflation situation, although this phenomenon is considered as a shield against inflation, but its secondary effect is that The companies are helpless when the debts are due and the reputation of the organization is damaged. Until now, various models have been used to predict financial helplessness. The patterns used in this field are very useful in the decisions of financial market actors. It has always been tried to improve the accuracy of prediction and evaluation of these patterns by using more advanced methods. The statistical population of this research includes all companies admitted to the Tehran Stock Exchange. The statistical sample includes 54 financially helpless companies and 54 healthy companies between the years 1990 and 1400, and in order to categorize the companies into the two mentioned groups, the default of Article 141 of the Commercial Law was used. The results show that the designed model has the ability to predict the occurrence of a financial crisis in companies admitted to the stock exchange up to two years before its occurrence. Also, the obtained results confirm the improvement of the prediction of helpless companies by entering the efficiency score into the models, but this improvement is not very significant.