1 Survival Data

Survival analysis is the statistics of time until something happens: death after a diagnosis, relapse after treatment, failure of a machine, the end of a spell of unemployment, the default of a loan. Such data look simple, a single positive number for each subject, but they have a feature that ordinary methods cannot handle: for many subjects the event has not happened by the time the study ends, so their times are only partly known. This section describes survival data, the kinds of incompleteness they have, and why that incompleteness changes the methods we must use.

Questions on this section

Stuck on something here? Ask below and it stays attached to this topic.