Competing Risks Models Using Mortgage Duration Data Under the Proportional Hazards Assumption

26 Pages Posted: 28 Mar 2012

See all articles by Mark Yuying An

Mark Yuying An

Federal National Mortgage Association (Fannie Mae); Duke University

Zhikun Qi

affiliation not provided to SSRN

Date Written: March 27, 2012

Abstract

This paper demonstrates two important results related to the estimation of a competing risks model under the proportional hazards assumption with grouped duration data, a model that has become the canonical model for the termination of mortgages with prepayment and default as two competing risks. First, the model with non-parametric baseline hazards is unidentifiable with only grouped mortgage duration data. Therefore, assumption on the functional form of the baseline hazard is necessary for any meaningful inference. Second, under some parametric assumptions such as piece-wise constant baseline hazards, the sample likelihood function has an explicit analytical form. Therefore, there is no need for the approximation formula widely adopted in the previous literature. Both Monte Carlo simulations and actual mortgage data are used to demonstrate the adverse impact of the approximation.

Keywords: proportional hazards, competing risks

Suggested Citation

An, Mark Yuying and Qi, Zhikun, Competing Risks Models Using Mortgage Duration Data Under the Proportional Hazards Assumption (March 27, 2012). Journal of Real Estate Research, Vol. 34, No. 1, 2012, Available at SSRN: https://ssrn.com/abstract=2029670

Mark Yuying An (Contact Author)

Federal National Mortgage Association (Fannie Mae) ( email )

3900 Wisconsin Avenue, NW
Mail Stop 1H3N-01
Washington, DC 20016-2892
United States
(202) 752-8442 (Phone)
(202) 752-5460 (Fax)

Duke University ( email )

Box 90097
207 Social Sciences
Durham, NC 27708-0097
United States
(919) 660-1809 (Phone)
(919) 684-8974 (Fax)

Zhikun Qi

affiliation not provided to SSRN ( email )

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
193
Abstract Views
1,150
Rank
283,739
PlumX Metrics