Logistic Robust Method to New Generalized Geometric Credit Risk Approach

This paper presents a complementary technique for empirical analysis of financial ratios and bankruptcy risk. Within this new framework, we propose the use of a new measure of risk, the Generalized Risk Box (GRB) measure. This method would be a general methodological guideline associated with financial data, including solving some methodological problems concerning financial ratios such as non-proportionality, non-asymmetry and non-scalability. In this paper, bankruptcy prediction and better accuracy rates obtained with GRB approach in compare to employing common ratios. This paper also suggests a Robust Logit method, which extends the Logit model by taking outlier into account. We employ Logit and Robust Logit Regression to assess our new method and sample forecast performances. Accuracy results show Robust Loigt method is substantially superior to the Logit method in financial studies