An Effective Meaningful Way to Evaluate Survival Models
This work addresses a practical problem in survival analysis for researchers and practitioners by providing a more reliable evaluation method, though it is incremental as it builds on existing metrics.
The paper tackles the challenge of evaluating survival prediction models with censored data by proposing a new metric based on pseudo-observations to estimate Mean Absolute Error (MAE), which accurately ranks models and often closely matches the true MAE, outperforming alternative methods.
One straightforward metric to evaluate a survival prediction model is based on the Mean Absolute Error (MAE) -- the average of the absolute difference between the time predicted by the model and the true event time, over all subjects. Unfortunately, this is challenging because, in practice, the test set includes (right) censored individuals, meaning we do not know when a censored individual actually experienced the event. In this paper, we explore various metrics to estimate MAE for survival datasets that include (many) censored individuals. Moreover, we introduce a novel and effective approach for generating realistic semi-synthetic survival datasets to facilitate the evaluation of metrics. Our findings, based on the analysis of the semi-synthetic datasets, reveal that our proposed metric (MAE using pseudo-observations) is able to rank models accurately based on their performance, and often closely matches the true MAE -- in particular, is better than several alternative methods.