Time-to-default analysis of mortgage portfolios

Issue title: Statistical Models in Finance and Insurance

Guest editors: Arkady Shemyakin and Vladimir Ladyzhets

Article type: Research Article

Authors: Galloway, M.^a | Johnson, A.^b | Shemyakin, A.^{c; *}

Affiliations: [a] , Minneapolis, MN, USA | [b] , St. Paul, MN, USA | [c] , St. Paul, MN, USA

Correspondence: [*] Corresponding author: A. Shemyakin, University of St. Thomas, St. Paul, MN, USA. E-mail: a9shemyakin@stthomas.edu.

Abstract: Fluctuation in mortgage default rates provides vital information to financial institutions and is a key indicator of the state of the economy. Using a decade’s worth (2002–2010) of data on prime and subprime mortgage portfolios, we propose and compare two models for mortgage defaults. The first, the Weibull-Gamma segmentation model (WGS), was utilized by Fader and Hardie (2007) in forecasting customer retention. Though effective in that setting, Markov chain Monte Carlo simulations suggest that the WGS suffers from over-parameterization. The Weibull segmentation model (WS) provides a simplified alternative that accurately forecasts default rates while identifying latent classes of “risky” prime and subprime mortgages characterized by increased hazard rates.

Keywords: Mortgage default, time-to-default, mixture model, latent class, Bayesian estimation

DOI: 10.3233/MAS-170409

Journal: Model Assisted Statistics and Applications, vol. 12, no. 4, pp. 359-367, 2017