Synthetic Dataset for Time-to-Event Analysis

该数据集是一个合成的数据集，旨在模拟观察性研究中治疗反应亚组的模型。在这个数据集中，基础的生存分布和分组结构是已知的。它包含了从正态分布中抽取的10个协变量，以及每种治疗制度下的生存分布，这些分布遵循戈珀茨分布。此外，该数据集模拟了随机化和观察性治疗分配情景。数据集的规模为30,000人，被划分为K=3个亚组。当前的任务是在观察性时间到事件数据中识别治疗反应亚组。

This synthetic dataset is designed to model treatment response subgroups in observational studies. The underlying survival distributions and subgroup structure of this dataset are fully known. It includes 10 covariates sampled from a normal distribution, alongside survival distributions for each treatment regimen that follow the Gompertz distribution. Additionally, this dataset simulates both randomized and observational treatment allocation scenarios. The dataset comprises 30,000 individuals and is partitioned into K=3 subgroups. The core task for this dataset is to identify treatment response subgroups from observational time-to-event data.