Dynamics of Modeling in Data Mining: Interpretive Approach to Bankruptcy Prediction 论文

摘要

Abstract:This paper uses a data-mining approach to develop bankruptcy prediction models suitable for normal and crisis economic conditions. It observes the dynamics of model change from normal to crisis conditions and provides interpretation of bankruptcy classifications. The bankruptcy prediction model revealed that the major variables in predicting bankruptcy were "cash flow to total assets" and "productivity of capital" under normal conditions and "cash flow to liabilities," "productivity of capital," and "fixed assets to stockholders equity and long-term liabilities" under crisis conditions. The accuracy rates of final prediction models in normal conditions and in crisis conditions were found to be 83.3 percent and 81.0 percent, respectively. When the normal model was applied in crisis situations, prediction accuracy dropped significantly in the case of bankruptcy classification (from 66.7 percent to 36.7 percent) to the level of a blind guess (35.71 percent). Therefore, the need for a different model in crisis economic conditions is justified.Key Words and Phrases: bankruptcy predictioncrisis managementdata miningdynamics of modeling Additional informationNotes on contributorsTae Kyung SungTae Kyung Sung is an Associate Professor of MIS at Kyonggi University, Korea. He received his Ph.D. in MIS from the University of Texas at Austin and a B.B.A. from Sungkyunkwan University. Dr. Sung's papers have been published in Technological Forecasting and Social Changes, Journal of MIS Research, Journal of Industrial Studies, Journal of Management Education and Research, Korean Management Science Review, Korean Management Review, Journal of Information Processing, International Business Review, and other journals. His research interests include information systems strategy, planning, and management, data mining and applications, business innovation, and knowledge/technology/information sharing and transfer.Namsik ChangNamsik Chang is an Assistant Professor at the College of Economics and Business Administration at the University of Seoul. He received his Ph.D. in MIS from the University of Arizona, an M.B.A. from the University of Missouri, St. Louis, and a B.S. from Korea University, Seoul. Dr. Chang has published in Decision Support Systems, International Journal of Information and Management Sciences, and Business Information Review, among other journals. He is currently conducting research in data-mining techniques and their applications to real-world domains, and database and data warehouse modeling.Gunhee LeeGunhee Lee is an Assistant Professor at the College of Business at Sogang University, Korea. He has a Ph.D. and an M.A. in statistics from the University of Missouri, Columbia, and a B.S. from Seoul National University. His publications have appeared in Environmental Toxicology and Chemistry and Journal of Korean Statistical Society. Dr. Lee's primary interests include data mining, reliability, statistical computing and modeling, Bayesian theory, and asymptotic theory.