Penalized likelihood estimation for rates with censored survival data
提出一种非参数方法，从删失生存数据中估计风险函数，通过惩罚似然得到非负双曲平滑样条估计，并利用计数过程的大样本理论建立渐近性质。
A non-parametric procedure is proposed for the estimation of the hazard function from censored survival data. The model is shown to fit into the Aalen multiplicative intensity model for counting processes and the estimator is derived by penalized likelihood methods. The estimate turns out to be a non-negative hyperbolic smoothing spline function. Standard largesample results for counting processes are used to establish asymptotic properties of the estimator and an example from censored survival analysis theory is presented.