Summary of Statistical methods for survival data Type

Summary of Statistical methods for survival data Type Nonparametric Advantages no assumption on theoretical distribution of survival time. Disadvantages Not flexible for multiple covariates; yields inefficient estimates at tail. Specific methods Kaplan. Meier Nelson. Aalen Life. Table Strong Semiparametric Parametric assumption The knowledge of the underlying distribution of survival times is not required. Flexible to handle multiple covariates (PH). Easy to interpret, more efficient and accurate when the survival times follow a particular distribution. When the distribution assumption is violated, it may be inconsistent and can give sub-optimal results. Cox model Regularized Cox. Boost Time-Dependent Cox Stratified Cox Tobit Buckley. James Penalized regression Accelerated Failure Time

Survival Analysis Methods in general Statistical Methods Kaplan-Meier Basic Cox-PH Lasso-Cox Non-Parametric Nelson-Aalen Penalized Cox Ridge-Cox Life-Table Time-Dependent Cox EN-Cox Regression Cox Boost Linear Regression Tobit Semi-Parametric Accelerated Failure Time Survival Trees Survival Analysis Methods Bayesian Methods Neural Network Machine Learning Support Vector Machine Buckley James Weighted Regression Panelized Regression Structured Regularization Naïve Bayesian Network Random Survival Forests Bagging Survival Trees Ensemble Active Learning Advanced Machine Transfer Learning Multi-Task Learning Informative censoring Related Topics Multistate model Complex Events Mixture Cure models OSCAR-Cox Competing Risks Multivariate data