Experimental Comparison of Semi-parametric, Parametric, and Machine Learning Models for Time-to-Event Analysis Through the Concordance Index
This work provides an empirical benchmark for survival analysis practitioners, but it is incremental as it compares existing methods without introducing new ones.
The paper compared semi-parametric, parametric, and machine learning models for time-to-event analysis using the concordance index on PBC and GBCSG2 datasets, finding that models with optimized hyperparameters generally outperformed those with default settings.
In this paper, we make an experimental comparison of semi-parametric (Cox proportional hazards model, Aalen's additive regression model), parametric (Weibull AFT model), and machine learning models (Random Survival Forest, Gradient Boosting with Cox Proportional Hazards Loss, DeepSurv) through the concordance index on two different datasets (PBC and GBCSG2). We present two comparisons: one with the default hyper-parameters of these models and one with the best hyper-parameters found by randomized search.