The objective of this case study is to perform Exploratory Data Analysis (EDA) on the given dataset to identify patterns whether a customer might default on Loan repayment based on a Consumer Finance Company dataset.
- The Case study entails doing EDA.
- We are provided with a single CSV file containing the Dataset from a Consumer Finance Company.
- Using this dataset, we need to find if we can establish some relationship between the various data points and the fact that customer might default or not.
- The majority of people (~74%) have taken a loan with a term of 36 months.
- As a general trend, we see that with the increase in years of experience, the number of loan seekers tend to decrease.
- Very few of the loan applicants own their own home.
- Loans that were "Charged Off" had a higher median interest rate of 13.59% as against a rate of 11.49% for loans that were fully paid.
- With increase in pub_rec_bankruptcies field, the percentage of people defaulting on loan payment rises.
- Median annual income of people who have defaulted on loan payment is lower than that of people who have not.
- If the number of 30+ days past-due incidences of delinquency in the borrower's credit file for the past 2 years is 7 or 8, there is a high risk of default, with the highest default rate at 8 delinquencies.
- In the highest grouping of total_acc field (field being grouped in steps of 10), the percentage of defaulting customers is the least.
- Pandas - version 2.1.4
- plotly - version 5.9.0
This project was based on the EDA Module of Course 1 (Statistics).
Created by [@rahul1092] - feel free to contact me!