Toward Conditional Distribution Calibration in Survival Prediction
This work addresses conditional calibration in survival prediction for real-world applications like individual decision-making, representing an incremental improvement over existing methods focused on marginal calibration.
The paper tackles the problem of improving conditional calibration in survival prediction models, which is important for individual decision-making, and demonstrates that their conformal prediction-based method effectively improves both marginal and conditional calibration without compromising discrimination across 15 diverse real-world datasets.
Survival prediction often involves estimating the time-to-event distribution from censored datasets. Previous approaches have focused on enhancing discrimination and marginal calibration. In this paper, we highlight the significance of conditional calibration for real-world applications -- especially its role in individual decision-making. We propose a method based on conformal prediction that uses the model's predicted individual survival probability at that instance's observed time. This method effectively improves the model's marginal and conditional calibration, without compromising discrimination. We provide asymptotic theoretical guarantees for both marginal and conditional calibration and test it extensively across 15 diverse real-world datasets, demonstrating the method's practical effectiveness and versatility in various settings.