top of page

TEAM

Predicting Credit Card Default

Song Gao, Juergen Kritschgau

clear.png

The goal of this project is to predict whether credit card defaults using basic demographic information and payment histories. The data is already clean with 23 variables that should be used to predict a binary output: will this person default next month; there are 30000 instances with no missing entries. This is a project from the "Possible Project Topics" google doc.
Upsides:
- clear business purpose
- clear measures of validation, i.e. we are predicting a binary outcome for which we have data
- we can compare results to an introductory paper (~89 accuracy using a random forest)
-dashboard possibilities: enter your payment history, to see if we think you will default
Downsides:
-the data is already so clean that we don't get to show off data cleaning skills

Data: https://archive.ics.uci.edu/dataset/350/default+of+credit+card+clients

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL
bottom of page