Credit Risk SQL Analysis

Project Overview

This project analyzes a credit risk dataset using SQL. The goal is to practice foundational SQL while answering business-style questions about loan defaults, borrower profiles, loan grades, interest rates, and loan purpose.

Dataset

The dataset used in this project is credit_risk_dataset.csv.

It contains:

• 32,581 records
• 12 columns
• borrower information such as age, income, home ownership, employment length, loan amount, loan interest rate, loan grade, loan purpose, default status, and credit history length

Table Name Used

For the SQL queries, the CSV should be imported into a table named:

credit_risk

If your database uses a different table name, replace credit_risk in the queries with your actual table name.

Columns Used

Column	Description
person_age	Borrower age
person_income	Borrower annual income
person_home_ownership	Borrower home ownership status
person_emp_length	Employment length in years
loan_intent	Purpose of the loan
loan_grade	Loan grade assigned to the loan
loan_amnt	Loan amount
loan_int_rate	Loan interest rate
loan_status	Loan status, where 1 indicates default and 0 indicates non-default
loan_percent_income	Loan amount as a percentage of income
cb_person_default_on_file	Whether the borrower has a previous default on file
cb_person_cred_hist_length	Credit history length in years

Business Questions

This analysis answers questions such as:

1. How many loans are in the portfolio?
2. What percentage of loans defaulted?
3. Which loan grades have the highest default rates?
4. Which loan purposes are more associated with default?
5. How does income differ between defaulted and non-defaulted loans?
6. How do home ownership and previous default history relate to loan status?
7. How can borrowers be grouped using simple CASE WHEN statements?
8. Are there missing values that should be reviewed before deeper analysis?

Key Dataset Checks

From the dataset review:

• Total records: 32,581
• Overall default rate: 21.82%
• Average income: $66,074.85
• Average loan amount: $9,589.37
• Average interest rate: 11.01%
• Missing values exist in person_emp_length and loan_int_rate, so these fields should be reviewed before deeper analysis.

SQL Skills Demonstrated

• SELECT
• WHERE
• GROUP BY
• ORDER BY
• COUNT, SUM, AVG, MIN, MAX
• simple percentage calculations
• simple CASE WHEN statements
• basic data quality checks

Project Structure

Credit-Risk-SQL-Analysis/
├── README.md
├── data/
│   └── credit_risk_dataset.csv
├── sql/
│   └── Credit_Risk_Analysis.sql

How to Use This Project

1. Create a table named credit_risk in your SQL database.
2. Import the file credit_risk_dataset.csv into that table.
3. Open sql/Credit_Risk_Analysis.sql.
4. Run each query one by one.
5. Review the comments above each query to understand the business question.

Notes

This project is not intended to be a predictive model. It is a SQL analysis project focused on understanding borrower and loan patterns related to default risk.

Possible Next Steps

Future improvements could include:

• creating a dashboard from the SQL outputs
• adding charts for default rate by loan grade and loan intent
• using Python to clean missing values and validate data quality
• expanding the project into a basic machine learning model after the SQL analysis is fully understood