자료포괄분석(DEA)을 이용한 신용평점모형의 개발

다국어 초록 (Multilingual Abstract)

Credit scoring problems are basically in the scope of classification problems that are commonly encountered decision making tasks in businesses. So far, a variety of methods such as linear probability and multivariate conditional probability models, recursive partitioning algorithm, artificial intelligence approach, multi-criteria decision-making (MCDM), and mathematical programming approach have been proposed to support the credit decision. Offering financial institutions a means for evaluating the risk of their credit portfolio in a timely manner, such models can provide an important body of information to help them formulate their respective risk management strategies. In fact, banking authorities such as Bank of International Settlements (BIS), the World Bank, the IMF, and the Federal Reserve all encourage commercial banks to develop internal models to better quantify financial risks.The purpose of this paper is to suggest a new approach to credit scoring, which is based on DEA. Compared with conventional methods such as multiple discriminant analysis (MDA), logistic regression analysis (LRA), and neural networks (NN) for business failure prediction, which require extra ex ante information of “good/bad”classification, this approach solely requires ex post information of the observed set of input and output data of the objects of interest (client firms) to calculate their respective credit scores. While NN and other traditional methods for credit scoring require ex ante information for business failure prediction, it is more useful in practice to build a credit scoring model based on ex post financial information. The idea is to develop a meaningful “peer group analysis”with specific financial characteristics that distinguish between two or more groups, and recently, data envelopment analysis (DEA) has been introduced to this peer group analysis for business failure prediction. DEA converts a multiplicity of input and output measures into a unit-free single performance index formed as a ratio of aggregated output to aggregated input. Conceptually, DEA compares the DMUs’observed inputs and outputs in order to identify the relative “best practices”for a chosen observation set. Based on these best observations, an empirical efficient frontier is established, and the degrees of efficiency of other units with respect to the efficient frontier are measured. Therefore, in the context of credit scoring, the performance index obtained via DEA can measure the relative credit riskiness of the firms within credit portfolio. In short, DEA, which computes a firm’s technical efficiency by measuring how well they transform their inputs into outputs relative to their peers, may provide a fine mechanism for deriving appropriate categories for credit scoring.For the empirical evidence, this DEA-based credit scoring methodology was applied to current financial data of external audited 1,061 manufacturing firms comprising the credit portfolio of the largest credit guarantee organization in Korea. Using financial ratios, the methodology could synthesize a firm’s overall performance into a single financial credibility score. The empirical results by this methodology were also validated by supporting analyses such as regression analysis and discriminant analysis, and by using actual bankruptcy cases. Comparing the results of this methodology against those obtained by regression analysis and discriminant analysis, and matching the results with actual bankruptcy cases of 103 firms, we could judiciously test the discriminatory power of this methodology.In addition, using the predicted financial credibility scores of client firms, we could provide a practical credit rating method to classify them into several balanced classes. Commercial banks or other financial institutions are currently adopting various credit rating methods to manage their respective client firms’credit riskiness, most of which utilize the probability of default derived by the neur