Decoding Small Business Administration (SBA) 7(a) Loan Charge Off Risk: A Predictive Model Odyssey

Machine learning (ML) and probability have been utilized by organizations such as the International Monetary Fund (IMF) to predict loan defaults for small and midsize enterprise (SME) loan defaults. Predominantly ML loan default models are based on loan application data and credit worthiness features. The goal of this praxis is to model US Small Business Administration (SBA) 7(a) loan charge off prediction utilizing loan application and economic features. The features are modeled with logistics regression, XG Boost, Decision Tree, Random Forest, and K-Nearest Neighbor ML methods. The model with the best precision for predicting if a loan will be paid in full or charged off is selected for a reusable model.

Author

• Franklin, Timalyn Eugenia

Language

• English

Keyword

• prediction
• small business
• loan
• default

Date created

• 2024

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Engineering Management

Degree

• D.Engr.

Advisor

• Jones, Michael
• Etemadi, Amir

Committee Member(s)

• Bierce, John

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/v979v390j