Distress Prediction and Stress Testing of Nonfinancial Firms: Case of Mongolia

This paper investigates the resilience of non-financial firms in Mongolia against financial distress. Utilizing firm-level financial data from 2013 to 2022, we employed a LASSO variable selection technique and logistic regression analysis to develop a distress prediction model for these firms. Among the 54 calculated financial ratios and indexes, the key indicators predictive of financial distress were identified as three profitability ratios, one liquidity ratio, one leverage ratio, and two financial indexes. Furthermore, our micro stress tests revealed that reductions in sales revenue significantly increase the likelihood of financial distress, with the probability rising to 32% under scenarios involving a 50% decline in sales. Additionally, sensitivity to income and expenditure shocks varies by firm size and economic sector. Firms in the mining and transportation sectors exhibit a higher probability of distress compared to those in the services sector. Similarly, micro and small firms are more vulnerable to distress than medium and large firms when subjected to stress scenarios.