ABSTRACT
This chapter focuses on the quantitative solution, through mathematical models, of the credit scoring problem, or classifying good or bad payers. The financial crises in recent years have highlighted the importance for moneylenders, banks, and regulators in assessing the credit risk. According to the Basel Committee, credit risk is the most serious problem in banking. Credit risk refers to the probability that a borrower fails to make the payments he owes. Classification problems such as the credit scoring issue can be faced with a variety of techniques, both parametric and nonparametric, as well as those that offer direct interpretation and those that do not. The Logit model enables a prediction of the probability whether or not an individual belongs to a group based on a series of independent variables that may be both quantitative and qualitative. Neurons carry out simple processes, transmitting their results to neighboring processors.
