Credit Default Modeling: a Logit Approach

Abstract

This paper aims at developing a credit scoring model that can best be used to ascertain the credit score and predict probability of default of firms seeking credit. The study subsequently aspires to find the financial ratios that can best be used to successfully construct the credit score and predict default risk. To achieve these purposes, the paper applied the logit model. Performance of the model was assessed using the percentage correctly classified (PCC) and the area under the receiver operating characteristic curve (AUC). The results show that the logit model yield very good performance rate for credit scoring and risk assessment. Further empirical evidence indicates that ratios bordering on: interest coverage, liquidity, activity, and firm size are those that can be significantly helpful in scoring credit applicants and assessing credit risk. Practically, the model can aid in reducing the time spent on evaluating credit applicants, and can give an exact default-risk intensity of each firm subjected to the model as well as serve as an early warning system. The multiplier effect will be a significant improvement in loan portfolio quality of the model user which is in accordance with the Basel II framework.

Keywords: Credit default modeling, logit, Ghana