Modeling Prepayments in Mortgage Backed Securities
Done at IMA at University of Minnesota, July, 2022
While participating at Math to Industry bootcamp, hosted by Institute for Mathematics and its Applications, University of Minnesota, during Summer 2022 we worked on a U.S. Bank sponsored project on predicting prepayments in Mortgage Backed Securities (MBS).
Investigated and created an 87% accurate linear regression model of various macroeconomic factors such as, Home Price Appreciation (HPA), Housing Credit Availability Index (HCAI), Geographic Mobility, on the rate of mortgage prepayments (CPR) in a top-down approach.
As a bottom-up approach assessed and modeled effect of current interest rate on prepayment rate among bor- rowers from 3 different cohorts of credit-score.
Due to the proprietary nature of the data used in this project we can not make this repository public. Please contact me over email if you are interested.