The AdviceRobo model is psychological statistical model, which it has been developing in close cooperation with Cambridge University. The model is based on borrowers’ motivations, personality traits, values, attitudes, interests, and lifestyles.
AdviceRobo developed, tested and validated its psychographic risk scoring model based on 2 million borrowers from multiple lenders across several countries throughout Europe, Latin America and Asia.
Most important risk metric is the Gini coefficient. In credit risk it is commonly used to evaluate the performance of the model. Gini assesses the accuracy of a prediction around whether a loan applicant will repay or default.
The proven Gini of the CreditRobo Logistic regression model is up to 0.66. With this, the AdviceRobo solution is the Best-in-class thin-file risk scoring solution. With our customers we have shown lifts of up to 49% in customers’ Gini’s.
On request, we will provide other risk metrics e.g. Confusion matrix, Precision, Recall, F1 or Accuracy.