Authors:

Danhe Tang

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

This project evaluated two predictive models to determine the most accurate model for banks to better predict the credit card attrition. Credit card attrition can have negative impacts on banks, causing revenue loss. It's a valuable capability for banks to predict credit card customer churn as it allows them to address customer concerns and retain business to boost the overall profit. I built and evaluated K-Nearest Neighbors (KNN) and Decision Tree models of 10,127 observations and select the most accurate model to forecast if the customer will churn based on misclassification rate. The model provides a way for bank practitioners to understand what attributes historically can be an indicator of credit card attrition and strategically identify the potential customer churn.

Source:

Purdue University / 2023

Topics:

No topics listed

Co-authors:

Danhe Tang