Contents

1. Jupyter notebook
2. Presentation
3. loan.csv.zip (Unzip the file to run the code)
4. README

Motivation

Analyse the loan application data and draw the recommendations to increase the fully paid loan applications and reduce the charged off loan applications.

Project Name

Lending Club Case Study.

Table of Contents

• General Info
• Technologies Used
• Conclusions

General Information

• Perform the data analysis on Lending Club loan applications to reveal the facts and make suggestions to minimize the charged off loan accounts.
• Identify the risk of loan application based on the past data analysis to help in deciding wether to approve or reject the loan application.

Conclusions

• State

  • Nebraska (NE) has highest percentage of charged off loans (60%)
  • Wyoming (WY) has lowest percentage of charged off loans (4.82%)

• Employment length

  • Applicant without any work experience more likely to default on loan (21.21%)
  • Applicant with 9 years of work experience more likely to replay loan (83.54%)

• Home Ownership

  • Applicants with home ownership status as “Other” more likely to default on loan (18.37%)
  • Applicants with “Mortgage” home status are more likely to replay loans (84.82%)

• Loan Purpose

  • Small Business loan purpose applications are more likely to default on loan (26%)
  • Wedding loan purpose are more likely to repay loans (89.84%)

• Loan Grades:

  • Percentage of charged off loan increases from Grade A to Grade G
  • Grade A loan applications are safest and the Grade G are riskiest.

Technologies Used

• numpy
• matplotlib
• seaborn
• numpy
• jupyter notebook

Contact

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