Small-Medium-Enterprise Financial Failure Prediction Considering Efficiency

With the rapid development of China's market economy, Small-Medium-Enterprises (SMEs) which develop with high speed recently, play an important role in national economy. Many growth SMEs have become the main force of market. However, SMEs meet many problems in the developing process. Financing is one of the bottlenecks. This is mainly because that SMEs are lack of perfect credit evaluation system and financial distress prediction system, which directly bring up two problems. One is that it is difficult for SMEs to make right decisions and scientific strategies about financial management and business management. Another is that, as creditors, banks and financial institutions are impossible to invest in SMEs because of lacking of warrants in estimating SMEs'financial status.Thus, the thesis tries to develop a feasible financial distress predicting method for SMEs,which considers SMEs'financing and operation characteristics and the low level of transparency of financial information systematically.The thesis firstly defines SMEs financial distress based on the introductions of the relative definitions about corporate financial distress, financial distress predicting indices and predicting methods. Then, by analyzing the business and financing characteristics of SMEs, this thesis presents an attribute system for SME financial distress prediction. Compared with the general attribute system of financial distress prediction, the main difference of the proposed attribute system is that the business efficiency is included. Business efficiency is evaluated by Data Envelopment Analysis (DEA). Finally the thesis develops a SME financial distress predicting method based on Support Vector Machine (SVM), and applies the method to the historical data which is collected from a financing support organization. Experimental results confirmed the rationality and validity of the proposed method and attribute system.