기업 재무자료를 활용한 설비건설기업 부실위험예측에 관한 연구

In this study, a mechanical equipment construction(MEC) industry insolvency risk prediction model was developed to predict the risk of insolvency of mechanical equipment construction companies which are small and medium-sized construction companies in advance and grade them to provide basic information on preparing countermeasures for industry stakeholders. For the analysis, the financial ratio calculated by the financial information of MEC companies subject to external audit from 2011 to 2021 was used. And the predictive performance between models was compared by applying three machine learning companies: Logit, Support Vector Machine(SVM), and Random Forest(RF). Corporate insolvency risk was set for those suffering from financial insolvency or marginal companies, and the model with the best model performance was set as the final predictive model by constructing a model reflecting a lag of 0 to 3 years between the dependent variable and the explanatory variable. As a result, it was confirmed that the zero-year lag logit model based on insolvency risk(financial insolvency) as a dependent variable was the most suitable for predicting insolvency risk of MEC companies. And based on this, the threshold and insolvency risk level of the classification of insolvent risk companies were calculated and presented. This study has a limitation in that it failed to reflect corporate know-how or macroeconomic conditions by setting up the prediction model using only financial information. But it is meaningful that a general model of predicting insolvency and reliable data from external audit companies were used to build and verify a insolvency risk prediction model for MEC companies, a small and medium-sized enterprise.