Modeling Default Risk for a Small Lender using Machine Learning

Individuals with limited or poor credit history are often unable to access credit from traditional sources such as banks. While some alternative lenders will provide credit to such individuals, these lenders typically lack reliable sources of information which can be used to accurately assess the risk that the loans they make will not be repaid, and thus tend to charge very high rates of interest to compensate for the uncertainty involved on such loans. This project aims to design a system that lenders can use to collect meaningful data about their customers, and then analyze this data for the purpose of making decisions about lending. By enabling the lender to make more accurate decisions about the risks entailed in their lending practices, this project can ultimately make it possible for lenders to offer more loans to customers with limited or poor credit history at more affordable rates.

Intern: 
Dalia Shuldiner
Faculty Supervisor: 
Brennan Thompson
Province: 
Ontario
University: 
Partner: 
Partner University: 
Discipline: 
Program: