Maham Taqi, a Master of Science candidate in the Department of Mathematics and Statistics, will virtually present the Masters Research Paper titled Lifetime Probability of Default Estimation for Consumer Loans under IFRS 9: A Comparative Study of Survival and Competing-Risks Models on Monday, July 20, 2026 at 10:00 AM.

The examination committee includes Supervisor Dr. Syed Ejaz Ahmed and Supervisory Committee Member Dr. Jan Vrbik.

Students (both graduate and undergraduate) as well as other members of the Brock Community are invited to attend. A Microsoft Teams link to the presentation can be found here: Join the meeting.  Please join with your camera and microphone turned off.

Keywords: IFRS 9; lifetime probability of default; survival analysis; competing risks; discrimination; calibration

Abstract: Lifetime probability of default (PD) estimation is an important component of expected credit loss measurement under IFRS 9. In consumer loan portfolios, modelling must account for both loan default and early prepayment over the contractual maturity. This study compares survival and competing-risk models for estimating lifetime PD in 36-month and 60-month loan cohorts. Model performance is evaluated using time-dependent AUC and Brier score, while calibration is assessed against the empirical Aalen–Johansen default cumulative incidence function. The research emphasizes the importance of evaluating lifetime credit-risk models across two contractual horizons, with particular attention to whether discrimination and calibration lead to the same model conclusions.  Discrimination is found to be similar across models, whereas calibration reveals meaningful differences between modelling approaches. These results suggest that calibration is more informative than discrimination, and that competing-risk methods provide a more appropriate framework for IFRS 9 lifetime PD estimation and expected credit loss measurement.