Financial Distress Prediction for Chinese Listed Manufacturing Companies

Financial distress prediction can be formulated as a classification problem and accomplished by advanced data mining techniques. In classification based on multiple criteria linear programming (MCLP), we need to find the optimal solution as a classifier, by solving the MOT problem. However, the errors can be caused by a fixed cutoff between a "good" group and a "bad" group by MCLP structure. In many applications, such as credit card account classification and bankruptcy prediction, how to handle two types of error is a key issue. Using the structure of multiple criteria and multiple constraint levels linear programming ((MCLP)-L-2), which allows alterable cutoff, two types of errors can be systematically corrected. In order to do so, a penalty is imposed to find the potential solution for all possible trade-offs in solving (MCLP)-L-2 problem. Real dataset of Chinese listed manufacturing companies is used to validate (MCLP)-L-2 method. Comparison with classical optimization-based method SVM and MCLP is also provided. (C) 2013 The Authors. Published by Elsevier B.V.