An Empirical Study Of Financial Distress Of Enterprise Groups In China Listed Companies

Enterprise Group occupies an important position in China's economic development and is the major commercial banks` credit customers. Its financial crisis will bring huge lose to creditors and other stakeholders. In order to enhance its ability to resist risk, it will be great significance for effectively monitoring and controlling the Group's financial risk by establishing an effective method of financial distress.This paper takes the Shanghai and Shenzhen A-share listed companies of enterprise groups in China as research object, analyzes the reason for their financial crisis forms, sets up financial distress indicator system, which is composed of 28 indicators, including five categories of basic financial indicators, character indicators of enterprise groups and corporate governance indicators. Then we divide the initial indicators to three parts for comparing the difference by adding the indicators of characteristics and corporate governance, the first one is including enterprise financial indicators, characteristics indicators and indicators of corporate governance; the second is including enterprise financial indicators and characteristics indicators; the last one is just including enterprise financial indicators. 70 companies from year 2004 to 2007 are selected as the training sample, of which 35 companies are"ST"enterprise groups to be listed and the rest are non-financial difficult enterprise groups to be listed. 15"ST"listed companies of enterprise groups and 15 non-financial difficult enterprise groups to be listed from year 2008 to 2010 are selected as the test sample. Then we carry on the factorial analysis for the three types of financial crises forecasting indicators to extract common factors, which are used to establish the financial crises forecasting model. Then we use logistic regression and support vector machine methods to build enterprise group of listed companies` financial distress prediction model and test.The results show that common factors in factor analysis have a large load on indicators of enterprise group characteristics and corporate governance, which includes information of reflecting the financial situation of enterprise groups, the introduction of these two types of indicators improve the effective and accuracy of early warning model. In addition, the models based on Logistic Regression and Support Vector Machine methods have better effect, and is feasible and effective. The model based on support vector machine method and including enterprise groups'characteristics and corporate governance indicators has the highest forecasting correct rate. It can reflect the enterprise group's financial crisis. When corporate managers or corporate stakeholders need to analyze the factors leading to the financial crisis while forecasting the financial crisis, using Logistic regression can visually analysis the variables` effect on the financial crisis.