Project Name

Lending Club Case Study

Table of Contents

• General Info
• Technologies Used
• Conclusions
• Acknowledgements

General Information

Lending Club is the largest online loan marketplace, offering personal, business, and medical procedure loans to urban customers through a fast online interface. When a loan application is received, the company must decide whether to approve it based on the applicant’s profile. Two types of risks are associated with this decision:

• losing business by not approving a loan to a likely customer
• financial loss by approving a loan to a likely defaulter.

The provided data includes information about past loan applicants and their default status. The aim is to identify patterns that indicate a likelihood of default, which can inform actions such as denying the loan, reducing the loan amount, or lending at a higher interest rate to risky applicants.

Business Objective

To understand the driving factors (or driver variables) behind loan default, i.e. the variables which are strong indicators of default. The company can utilise this knowledge for its portfolio and risk assessment.

Conclusions

** Summary - Results **

• 86% increase in last year from 2010 to 2011, with more short-term loans for all top products apart from Small Business
• Debt Consolidation Loan most popular with 47.2%.
• Overall default rate is 14%, which constitutes B grade and short-term loans more
• Default rate is highest for Debt Consolidation loan and for customers with 10+ years employment length
• People with Rent and Mortgage constitute the highest number of defaulters, along with people with income band 40k-60k
• People with no past delinquency and no derogatory public records are more likely to default
• Loans which are fully funded are most likely to default, compared to loans that are less than 50% funded

** Recommendations **

• Avoid B, C and D category loans in favor of A category loans
• Favor Homeowners more in lieu of those with Rent and mortgage
• Disburse loans for customers with shorter than 10 years of employment length
• Fully Funded loans need to be avoided by allowing up to 60% of overall ask
• People with delinquency and / or past derogatory public record should not be ignored for loan

Technologies Used

• pandas==2.2.2
• numpy==1.26.4
• matplotlib==3.9.1
• seaborn==0.13.2

Acknowledgements

Contact

Created by [@sjpathak] - feel free to contact me!