Forecasting and stress testing credit card default using dynamic models

Tony Bellotti, Jonathan Crook

Research output: Journal PublicationArticlepeer-review

73 Citations (Scopus)


We present discrete time survival models of borrower default for credit cards that include behavioural data about credit card holders and macroeconomic conditions across the credit card lifetime. We find that dynamic models which include these behavioural and macroeconomic variables provide statistically significant improvements in model fit, which translate into better forecasts of default at both account and portfolio levels when applied to an out-of-sample data set. By simulating extreme economic conditions, we show how these models can be used to stress test credit card portfolios.

Original languageEnglish
Pages (from-to)563-574
Number of pages12
JournalInternational Journal of Forecasting
Issue number4
Publication statusPublished - Oct 2013
Externally publishedYes


  • Credit risk
  • Stress testing
  • Survival analysis

ASJC Scopus subject areas

  • Business and International Management


Dive into the research topics of 'Forecasting and stress testing credit card default using dynamic models'. Together they form a unique fingerprint.

Cite this