[link]
The paper looks at approaches to predicting individual survival time distributions (isd). The motivation is shown in the figure below. Between two patients the survival time varies greatly so we should be able to predict a distribution like the red curve. https://i.imgur.com/2r9JvUp.png The paper studies the following methods:  classbased survival curves KaplanMeier [31]  KalbfleischPrentice extension of the Cox (coxkp) [29]  Accelerated Failure Time (aft) model [29]  Random Survival Forest model with KaplanMeier extensions (rsfkm)  elastic net Cox (coxenkp) [55]  Multitask Logistic Regression (mtlr) [57] Looking at the predictions of these methods side by side we can observe some systematic differences between the methods: https://i.imgur.com/vJoCL4a.png The paper presents a "DCalibration" metric (distributional calibration) which represents of the method answers this question: Should the patient believe the predictions implied by the survival curve? https://i.imgur.com/MX8CbZ7.png
Your comment:
