Incorporating data drift to perform survival analysis on credit risk

2026-01-28

AI Summary

This study addresses the limitations of traditional credit risk survival models, which assume data stationarity and thus struggle to adapt to distributional shifts caused by changes in borrower behavior, macroeconomic conditions, or regulatory policies. To overcome this, the authors propose the first dynamic joint framework that explicitly models data drift mechanisms by integrating longitudinal account balance trajectories with a discrete-time hazard model. The approach unifies the handling of abrupt, gradual, and periodic drifts through landmark one-hot encoding and isotonic recalibration. Empirical evaluation on Freddie Mac mortgage data demonstrates that the proposed method consistently outperforms classical survival models, tree-based adaptive techniques, and gradient boosting algorithms across diverse drift scenarios, achieving superior discrimination and calibration performance.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Mixed Discrete/Continuous Search
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Survival analysis has become a standard approach for modelling time to default by time-varying covariates in credit risk. Unlike most existing methods that implicitly assume a stationary data-generating process, in practise, mortgage portfolios are exposed to various forms of data drift caused by changing borrower behaviour, macroeconomic conditions, policy regimes and so on. This study investigates the impact of data drift on survival-based credit risk models and proposes a dynamic joint modelling framework to improve robustness under non-stationary environments. The proposed model integrates a longitudinal behavioural marker derived from balance dynamics with a discrete-time hazard formulation, combined with landmark one-hot encoding and isotonic calibration. Three types of data drift (sudden, incremental and recurring) are simulated and analysed on mortgage loan datasets from Freddie Mac. Experiments and corresponding evidence show that the proposed landmark-based joint model consistently outperforms classical survival models, tree-based drift-adaptive learners and gradient boosting methods in terms of discrimination and calibration across all drift scenarios, which confirms the superiority of our model design.

Problem

• data drift
• survival analysis
• credit risk
• non-stationary environment
• model robustness

Innovation

• data drift
• survival analysis
• dynamic joint modelling
• landmark encoding
• credit risk