Optimal Survival Trees
This work addresses the problem of improving the accuracy of survival tree models for researchers and practitioners working with censored outcomes, particularly in medical data.
This paper introduces the Optimal Survival Trees (OST) algorithm, which utilizes mixed-integer optimization and local search to create globally optimized survival tree models. The OST algorithm demonstrates improved accuracy compared to existing survival tree methods, especially when applied to large datasets.
Tree-based models are increasingly popular due to their ability to identify complex relationships that are beyond the scope of parametric models. Survival tree methods adapt these models to allow for the analysis of censored outcomes, which often appear in medical data. We present a new Optimal Survival Trees algorithm that leverages mixed-integer optimization (MIO) and local search techniques to generate globally optimized survival tree models. We demonstrate that the OST algorithm improves on the accuracy of existing survival tree methods, particularly in large datasets.