Survival analysis deals with time to an event in systems. Events can be death in biological systems and failure in technical systems, but the event may be something entirely different, such as divorce, relapse of a disease, or an insurance claim. Often the time to an event is not known exactly, but is known to fall in some interval; this is called censoring. The Wolfram Language allows you to specify time-to-event data in a flexible (censor intervals, indicators, or counts) and powerful (right, left, interval censoring, and truncation) way. Time-to-event data is broadly supported throughout the system. Time-to-event data can be used to compute descriptive statistics, estimate parametric and nonparametric distributions, fit a variety of survival models, and perform hypothesis tests.
EventData — right, left, interval-censored, and truncated data
Median — median life of survival data and distributions
SurvivalModelFit — distribution with confidence intervals (Kaplan–Meier, …)
EstimatedDistribution — estimate parametric distribution from survival data
CoxModelFit — estimate a Cox proportional hazards model
LogRankTest — test whether hazard rates are equivalent