An Application of Machine Learning to Mortgage Prepayment Modeling

The business partner is interested in expanding its understanding of prepayment. Specifically the goal is to predict prepayment risk for mortgages as a function of mortgagors’ characteristics (including data from previous interactions with the bank), and the local economy. In recent years, with the improvement in efficient computing and data storage, the relying on a wide range of mortgagors’ characteristics to predict prepayment risk has become more prevalent in the industry. Given the large amount of data collected by financial institutions for mortgages, extracting information that can be used to predict prepayments requires appropriate statistical techniques to perform big-data analysis

Faculty Supervisor:

Valentina Galvani;Sebastian Fossati Pereira

Student:

Partner:

ATB Financial

Discipline:

Business

Sector:

Finance and Insurance

University:

University of Alberta

Program: