Doubly Robust Conformalized Survival Analysis with Right-Censored Data
This work addresses survival analysis for medical or reliability applications, offering a robust method for right-censored data, though it is incremental as it builds on existing conformal inference approaches.
The paper tackles the problem of constructing lower prediction bounds for survival times from right-censored data by extending conformal inference methods, achieving robust predictive inferences with theoretical support and empirical validation in challenging settings.
We present a conformal inference method for constructing lower prediction bounds for survival times from right-censored data, extending recent approaches designed for more restrictive type-I censoring scenarios. The proposed method imputes unobserved censoring times using a machine learning model, and then analyzes the imputed data using a survival model calibrated via weighted conformal inference. This approach is theoretically supported by an asymptotic double robustness property. Empirical studies on simulated and real data demonstrate that our method leads to relatively informative predictive inferences and is especially robust in challenging settings where the survival model may be inaccurate.