Credit risk is defined as the probability of a loss when the borrower can not pay back a loan or the borrowed sum. If customers don’t repay their credit, the lender loses money. And, if this loss happens on a large enough scale, it’s impact can be quite huge.
Most common scoring tools that we find in the wild like the Fico or the Vantage scores, only takes into account financial factors and previous credit performance of the customers.
We want to test a model that allows us to take into account the profiling of the customers. Factors like where they life, for how long have they been working, how many children do they have, are they single or married, what is their education background or do they own a car or house?
- Agree on a definition of GOOD/BAD client based on credit repay record
- Analyze the factors that affect the determination of GOOD/BAD clients
- Build a model that classifies/predicts the customers as GOOD/BAD for credit approval