Competing risks of default and prepayment of mortgage market

File Type:
PDFItem Type:
ThesisDate:
2020Author:
Access:
openAccessCitation:
OLAJUBU, OLUWATOBILOBA JOHN JOHN, Competing risks of default and prepayment of mortgage market, Trinity College Dublin.School of Computer Science & Statistics, 2020Download Item:
Abstract:
Using a large data set on the Single family home loans from The Federal Home Loan Mortgage Corporation (FHLMC), sponsored by the US government, this research studies the economic factors affecting the competing risks of prepayment and default. Since the subprime crisis,understanding the competing risks of prepayment and default on the single-family mortgage market have become increasingly important. Refinancing incentives on the mortgage market has been triggered by the current low interest rate environment. The paper presents a unified model of the competing risks of default and default mortgage termination, considering the two hazards as independent competing risks, which are jointly estimated. It also accounts for the unobserved heterogeneity among borrowers, and estimates the unobserved heterogeneity simultaneously with the parameters and baseline hazards associated with default and prepayment functions. This thesis also examines the different models that can be used to predict current and future prepayment rates. In mortgage research applications the time to occurrence of one event of interest which may be caused by another competing event is investigated. The most widely used prepayment and default models are competing risk models and option theoretic models. However, this thesis places more emphasis on competing risk models. As we know, survival analysis examines and models the time to an event (usually failure of a system and possibly subject to right censoring) and one of the most common tools 'Cox proportional-hazards regression model' is used to study the dependency of survival times on the predictor variables. Important risks determinants include the specific characteristics of the borrower, specific characteristics of the loan and macro-economic variables. Variable selection procedures are used to identify the most significant risk drivers. Tests on parameter stability are conducted to check forecasting accuracy for models. A problem, however, is heterogeneity between mortgages-suggesting there is future potential research to be done on this topic.
Author's Homepage:
https://tcdlocalportal.tcd.ie/pls/EnterApex/f?p=800:71:0::::P71_USERNAME:OLAJUBUODescription:
APPROVED
Author: OLAJUBU, OLUWATOBILOBA JOHN JOHN
Advisor:
Wilson, SimonPublisher:
Trinity College Dublin. School of Computer Science & Statistics. Discipline of StatisticsType of material:
ThesisAvailability:
Full text availableKeywords:
Cox Proportional-hazards model, Competing risks, Mortgages, Default, Pre-paymentMetadata
Show full item recordLicences: