Performance of Minority Data in Financial Distress Prediction Models. Application of Multiple Discriminate Analysis, Logit, Probit and Artificial Neural Network Methods
Performance of Minority Data in Financial Distress Prediction Models. Application of Multiple Discriminate Analysis, Logit, Probit and Artificial Neural Network Methods

Summary/Abstract: This paper presents a financial distress prediction model that combines the approaches of multiple discriminant analysis, logit method, Probit method and artificial neural network for distinguishing between healthy and financially distressed firms. Financial distress prediction is a key issue in finance. Past research has largely ignored the precision of the minority group and focused on the overall accuracy percentage. This paper suggests artificial neural network is the best high performance method for both the groups. The suggested method serves as an earlywarning system for firms in financial distress.

• Details
• Contents