Logistic Model Of Corporate Financial Distress Prediction Based On Distance-to-Default

With the constant expansion of global equity markets and the further openness of China's financial industry, there are opportunities as well as unprecedented challenges facing listed companies of China. Banks, corporate owners and other market players are in an urgent need of an effective early-warning mechanism to help in their decision-makings.The current financial early-warning models can be divided into two categories:the accounting ratio models and the credit risk quantitative models respectively represented by logistic model and KMV model. Despite its simplicity and flexibility, the former model is too hysteretic to capture the tiny and rapid changes of the capital market. For the latter model, given that it's based on the market information rather than the historical data to predict, it can, theoretically, reflect a company's status of finance and operation in a scientific, dynamic and timely way. However, due to lack of credit database, the ultimate output of KMV model can not be produced in China.Defining specially treated (ST) companies as financially distressed companies, this article selects 225 manufacturing enterprises listed in Shanghai &Shenzhen A-share stock markets and respectively establishes three models with different types of predictors involved:ModelⅠis based on the conventional accounting ratios, while ModelⅡis based on the ModelⅠratios with some non-financial predictors added,and ModelⅢis further modified by including both distance-to-default and the conventional indicators. Finally, we will use both classification matrix and ROC curve to test and compare the predictive power and goodness of fit of the three models.The empirical results show that the modified model with distance-to-default outperforms the conventional ones both in its predictive capability and its goodness of fit.