A method for forecasting customer behavior. Using data on a specific person, other people with the same characteristics are grouped together. In this way, the payment behaviour and risk of the group are calculated.
There are various rating classes in the credit rating scale. Points are assigned on the basis of borrower characteristics such as place of residence, occupation, collateral, etc., and these are then combined into a single credit rating grade. This then makes it easier to decide whether to grant a loan. If the creditworthiness is sufficient, a loan can be granted.
Rules and algorithms of scoring are called "scorecard". Techniques to develop scorecards are logistic regression, discriminant analysis, artificial neural networks and other data mining methods.