A new discrimination measure for assessing predictive performance of non-linear survival models
By: Alfensi Faruk, Jan Palczewski, Georgios Aivaliotis
Potential Business Impact:
Measures how well models predict survival times.
Non-linear survival models are flexible models in which the proportional hazard assumption is not required. This poses difficulties in their evaluation. We introduce a new discrimination measure, time-dependent Uno's C-index, to assess the discrimination performance of non-linear survival models. This is an unbiased version of Antolini's time-dependent concordance. We prove convergence of both measures employing Nolan and Pollard's results on U-statistics. We explore the relationship between these measures and, in particular, the bias of Antolini's concordance in the presence of censoring using simulated data. We demonstrate the value of time-dependent Uno's C-index for the evaluation of models trained on censored real data and for model tuning.
Similar Papers
Stop Chasing the C-index: This Is How We Should Evaluate Our Survival Models
Methodology
Fixes how we check if predictions of when things happen are good.
The C-index Multiverse
Machine Learning (Stat)
Makes computer predictions of sickness more trustworthy.
Beyond Cox Models: Assessing the Performance of Machine-Learning Methods in Non-Proportional Hazards and Non-Linear Survival Analysis
Machine Learning (CS)
Finds better ways to predict when people will get sick.