This case study aims to identify patterns which indicate if a client has difficulty paying their installments which may be used for taking actions such as denying the loan, reducing the amount of loan, lending (to risky applicants) at a higher interest rate, etc. This will ensure that the consumers capable of repaying the loan are not rejected. Identification of such applicants using EDA is the aim of this case study.
In other words, the company wants to understand the driving factors (or driver variables) behind loan default, i.e. the variables which are strong indicators of default. The company can utilize this knowledge for its portfolio and risk assessment.
-
The impact of education level Using a bar chart, we can see that Highest Risk Group: Individuals with only a Lower secondary education exhibit the highest default rate, peaking at approximately 11%. This suggests that lower educational attainment may be a significant risk factor in credit scoring.
Medium Risk Group: The categories of Secondary / secondary special and Incomplete higher show similar default behaviors, with rates hovering between 8.5% and 9%.
Lowest Risk Group: Candidates with Higher education show a significantly improved credit profile with a default rate of roughly 5.3%. Those with an Academic degree represent the lowest risk segment, with a default rate of less than 2%.
-
Impact of age Using a kde plot, we can see that, Younger Demographic Risk: There is a significantly higher density of Target 1 (defaulters) among younger clients, particularly those between the ages of 25 and 40. The peak for defaults occurs around age 30.
Older Demographic Stability: Conversely, as age increases (beyond 45 years), the density of Target 0 (repayers) becomes noticeably higher than that of Target 1. This suggests that older clients are statistically more likely to repay their loans.
Intersection Point: The two distributions cross at approximately 42–45 years of age. Beyond this point, the likelihood of being a "safe" borrower increases steadily.
-
Impact of external credit Using a kde plot, we can see that, Predictive Power: The lack of total overlap between the blue and orange areas indicates that external credit is a powerful predictor of default. Lower scores are highly indicative of potential repayment issues.
Critical Threshold: The intersection of the two curves occurs near the 0.45 mark. Clients scoring above this threshold have a significantly higher probability of being in the "repaid" category compared to those scoring below it.
-
Impact of previous application status Using a bar chart, we can see that, Impact of Prior Refusals: Clients whose previous applications were Refused show the highest current default rate, reaching 12%.
Historical Success as a Predictor: Clients with previously Approved applications have the lowest current default rate, falling below the Overall Average Default line (approximately 8%).
Intermediate Risk: Those with Canceled or Unused offers fall in the middle, though both groups still exhibit default rates near or above the population average.