Researchers have developed a new dynamic joint modeling framework to improve the robustness of credit risk survival analysis under non-stationary environments. This framework integrates a longitudinal behavioral marker with a discrete-time hazard formulation and landmark one-hot encoding. Experiments on Freddie Mac mortgage loan datasets demonstrated that the proposed model consistently outperformed classical survival models and other adaptive learners across various data drift scenarios, confirming its superiority in discrimination and calibration. AI
IMPACT This research offers a more robust method for credit risk assessment in financial institutions by addressing data drift, potentially leading to more accurate default predictions.
RANK_REASON Academic paper on a statistical modeling technique. [lever_c_demoted from research: ic=1 ai=0.4]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
