Prediction of Survival Probability Based on a Linear Regression Model
本文提出简单方法，利用线性回归模型预测患者的t年生存概率或中位失效时间，无需复杂非参数估计，并以斯坦福心脏移植数据为例说明。
After fitting survival data with a linear regression model, it is important to know how to use the results to make prediction of the t-year survival probability or median failure time for future patients with certain covariates. We propose some simple prediction procedures without using complicated and unstable nonparametric functional estimates. The new methods are illustrated with the Stanford heart transplant data.