UCI German Credit Data - EDA & XGBoost

Desciption & Background:

Credit scoring is a technique used by lenders and financial institutions to assess the creditworthiness of potential borrowers. It involves evaluating an individual's credit history and financial behavior to predict the likelihood of them defaulting (not paying back) on loan payments.

A higher credit score indicates a lower risk of default, while a lower credit score indicates a higher risk. Credit scoring is widely used in lending decisions, such as approving or denying loan applications, setting interest rates, and determining credit limits. It helps lenders make informed decisions, manage risk. 

For this project, I did a simple exploratory data analysis and fit a gradient-boosted decision tree (GBDT) machine learning model, XGBoost.

Keywords: Credit Risk, EDA, Gradient-Boosting, Classification

Result Summary:

• Non-Default Recall: 72%

• Default Recall: 65.6%

• AUC: 0.756