企業의 不實豫測模形에 관한 硏究 = A Study on Models in Business Failure Prediction

다국어 초록 (Multilingual Abstract)

Early warning system for business failure prediction will be helpful to many interest groups of a firm. Business failure prdiction is a case of binary classification in which the number of groups is limited to two. Numerous methods have been developed for classification purpose. They are typically referred to a multivariate discriminant analysis(MDA).
Thereafter, Logit analysis and Probit analysis are developed for failure prediction. These models are not deterministics but stochastic. In the current study, neural network approach is developed as a tool to discriminate business failure.
The purpose of this paper is to review the models in business failure prediction and verify the posibilities and limitations of neural net model as a classification tool.